Relationships between Flower Opening Time, Environmental Conditions, Corolla Opening Size and Nectar Production in Five Winter Oilseed Rape (Brassica napus L.) Cultivars in China
, , , , , y | 30 jun 2023
Article Category: Original Article
Publicado en línea: 30 jun 2023
Páginas: 57 - 72
Recibido: 02 ene 2023
Aceptado: 04 abr 2023
DOI: https://doi.org/10.2478/jas-2023-0005
Palabras clave
© 2023 Ran Liu et al., published by Sciendo
This work is licensed under the Creative Commons Attribution 4.0 International License.
Oilseed rape (
Researchers have observed and analysed the nectary structure and nectar production in different oilseed rape cultivars (Davis et al., 1996). Nectar is synthesized and secreted by the nectary, a special organ of entomophilous plants. Each flower bears two pairs of nectaries, of which two are inner nectaries and two are outer nectaries. The nectar secretion of oilseed rape flowers changes gradually over the course of the transition from bud dehiscence to gradual opening, blooming and then closing (Chabert et al., 2018). Nectar flow begins at the stage in which the yellow corolla is fully expanded. Initially, only the inner nectaries secrete nectar, and then the outer nectaries begin secretion when the four long stamens curve slightly towards the flower centre (Eisikowitch, 1981). When stigma receptivity decreases and oilseed rape flowers begin to senesce, nectar secretion capacity decreases (Chabert et al., 2018).
Nectar production, synthesis and secretion are linked to genetics and the environment (Liu & Thornburg, 2012; Bodó et al., 2021). Different cultivars of oilseed rape produce different nectar volumes, with various sugar contents and glucose/fructose ratios (G/F ratios) (Picard-Nizou et al., 1995; Pierre et al., 1999; Carruthers et al., 2017). Oilseed rape nectar is very low in sucrose and rich in glucose. Only hexose has been detected in this nectar, and most oilseed rape cultivars produced nectar with G/F ratios exceeding 0.95 (Kevan et al., 1991; Farkas, 2008). The external environment also has a great influence on nectar secretion (Enkegaard et al., 2016). For example, nectar quantity and concentration have been found to be correlated with air temperature and relative humidity (Mohr & Jay, 1990). The amount of nectar sugar per flower depends on the relative rates of secretion and reabsorption, and the sugar concentration of nectar is related to the ambient humidity (Burquez & Corbet, 1991). When the moisture content in the air is too low, the water in the nectar easily evaporates, leading to an increase in nectar sugar concentration (Corbet et al., 1979; Bertsch, 1983; Nicolson, 1998). In recent years, the phenomenon of global warming has gradually intensified, and the impact of rising air temperatures on nectar secretion has received widespread attention (Hassan et al., 2017; Takkis et al., 2018). Moderate increases in air temperature are conducive to the secretion of nectar, but extreme increases in air temperature can have a negative impact on nectar secretion (Petanidou & Smets, 1996; Takkis et al., 2015).
To better understand the nectar secretion patterns and for further assessing of the melliferous potential of oilseed rape cultivars, we carried out the present study under field conditions. For five winter oilseed rape cultivars, first, we assessed nectar secretion at the flower bud stage. Then, we analysed and compared the nectar secretion dynamics in flowers opening at different times throughout the day. The nectar sugar composition was also determined and compared among cultivars. Finally, the correlations of corolla opening size, air temperature and relative humidity with nectar volume and nectar sugar concentration were determined. The results will provide helpful information for a better understanding of the temporal dynamics of plant-pollinator interactions.
This study was carried out in Luoping County, Yunnan Province (24°53′ N; 104°20′ E), from February 25th to March 12th, 2021. The local average annual air temperature is 15.1°C, and the precipitation is approximately 1700 mm. The study site was a 0.35-ha field sown with five winter oilseed rape (
During the flowering period, we randomly marked 6–12 flower buds from 3–4 plants of each cultivar every three days, and we repeated this procedure five times. For each cultivar, a total of 42 flower buds were marked and observed. Insect proof meshes were used to bag these buds to prevent visiting insects from collecting nectar and affecting the nectar secretion quantity. The flower buds were bagged at 18:00 in the evening or at 9:00 in the morning. To assess the diurnal nectar secretion dynamics, we collected nectar from individual bagged flowers at 2-h intervals during the daytime (from 9 a.m. to 5 p.m.) and the flowers were re-bagged between nectar extractions. The flower buds bagged at 18:00 were monitored from 9:00 the next day, and the flower buds bagged at 9:00 were monitored from 11:00 on the same day until they began to wilt.
The nectar was collected using a microcapillary tube with a length of 100 mm and an inner diameter of 0.3 mm (Mesquida et al., 1988). Nectar volume, expressed in microliters, was calculated through the multiplication of the basal area and the length of the glass micropipette filled with nectar. The concentration of total sugar in the nectar (percentage of mass sugar/total mass solution) was measured with the use of a hand-held refractometer (0.0–53.0%, wt./total wt.; MASTER-53α, Atago, Japan). In addition, we randomly collected 10–15 flower buds from at least six different plants of each cultivar and assessed the nectar secretion volume and the nectar sugar concentration.
We determined the nectar volume and nectar sugar concentration per flower at each sampling time point throughout the flower lifetime. The daily nectar volume per flower was the sum of the nectar volume collected on that day, and the daily nectar sugar concentration was the average of several nectar sugar concentrations measured on that day. The total nectar yield per flower was the total quantity of nectar collected throughout the flower lifetime, and the average nectar sugar concentration was the average of the nectar sugar concentrations measured during the flower lifetime.
From when flowers opened until they began to wither, we also measured the corolla opening size (the maximum distance between the edges of two opposite petals) (N=42 for each cultivar) at each time point when the nectar was collected. During the experiment, the real-time air temperature and relative humidity from 9:00 to 18:00 were also recorded with the use of a mini data logger (RC-4HC, Elitech, China).
During the flower blooming period, nectar was collected from 1–50 fully open flowers late in the morning to fill the clean glass capillary micropipettes to obtain approximately 3.533 µL of nectar per sample. The collected nectar was expelled from the pipettes onto filter paper and allowed to air dry, following the protocol of Kevan et al. (1991) and Davis et al. (1996). Each piece of dried filter paper was placed in a 5 mL centrifuge tube and taken back to the laboratory. Six duplicate biological samples were collected from at least six different plants per cultivar.
The sugar composition of the nectar was studied according to the methods of Sun et al. (2017). The nectar spot was cut from the filter paper and eluted with 0.4 mL distilled water. The sugar content was determined by high-performance liquid chromatography (HPLC) conducted on a Cosmosil Sugar column (Sugar D, 5 µm, 250 mm×4.6 mm i.d.) with a differential refractive index detector (Waters 2695). We injected 15 µL of each sample into the column. The temperature of the column was maintained at 40°C, and the solvent was an acetonitrile: water system (75:25 by volume) at a flow rate of 1 mL/min. Standard solutions of glucose, fructose and sucrose were analysed to identify the specific sugars and calculate their relative amounts in each sample.
The Shapiro-Wilk normality test and Levene’s test were used to test the normality and homoscedasticity of the data. The differences in nectar volume, nectar sugar concentration and sugar composition were analysed with the use of one-way ANOVA or an independent samples t test when the data met the normality and homoscedasticity requirements. The nonparametric Kruskal-Wallis test or Mann-Whitney U test was used when the data did not meet the parametric assumptions. Generalized linear mixed models were used to analyse the interaction between cultivar and time of day for nectar volume and nectar sugar concentration, the interaction between cultivar and flowering day for daily nectar volume and daily nectar sugar concentration, and the interaction between cultivar and flowering opening time for the total nectar yield and the average nectar sugar concentration. The correlations of corolla opening size with nectar volume and nectar sugar concentration were analysed with the use of the Pearson correlation analysis. Stepwise regression analysis was performed to explore the relationship between nectar secretion and weather condition parameters. The multicollinearity was detected through variance inflation factor (VIF) values. Two-tailed probabilities for Fisher’s exact test were used to compare the percentage of flower buds that secreted nectar among the cultivars. All analyses were performed with Statistical Product and Service Solutions (SPSS) v27.0 (Morgan et al., 2004). Data are presented as means±standard errors.
The analyses showed that 30–75% of the flower buds secreted nectar. No significant differences were detected in the percentage of nectar-secreting flower buds among the five cultivars (Fisher’s exact test, p=0.162) (Tab. 1). The nectar volume secreted per bud varied between 0.247±0.020 μL and 0.724±0.355 μL, and there was no significant difference among cultivars (Kruskal-Wallis test: H=4.717, df=4, p=0.318). The nectar sugar concentration of ‘Dehuiyou’ (7.7±2.3%) was significantly lower than that of ‘Hengheyou 998’ (15.5±1.5%) (p=0.022). There were no significant differences in the nectar sugar concentration among the other four cultivars (p>0.05) (Tab. 1).
Nectar secretion volume (μL) and nectar sugar concentration (%, wt./total wt.) in the five oilseed rape cultivars at the bud stage
| 12 | 33.3a | 0.724±0.355a | 11.8±0.4ab | |
| 12 | 75.0a | 0.330±0.169a | 10.3±0.9ab | |
| 11 | 63.6a | 0.252±0.092a | 7.7±2.3b | |
| 10 | 30.0a | 0.247±0.020a | 15.5±1.5a | |
| 15 | 53.3a | 0.353±0.115a | 10.5±0.4ab |
Different letters in columns indicate significant differences (p<0.05).
From flower opening to withering, the corolla opening size of the five cultivars ranged from 0–31 mm (Fig. 1A & 1B). The flower buds that were bagged at 18:00 opened before 9:00 on the next morning and closed at approximately 17:00 on the second day after bagging. On the first flowering day, the corolla opening size gradually increased from 15–21 mm at 9:00 and reached the maximum mean value of 26 mm during the period from 13:00–17:00. By the second day, the corolla opening size had dramatically decreased, and the petals began to close and wither until 17:00 (Fig. 1A). For the flower buds that were bagged at 9:00 and that opened at approximately 11:00, the lifespan of a flower was approximately 50 hours. The corolla opening size showed an increasing trend on the first and second flowering days and reached a maximum mean value ranging from 24–26 mm from 13:00–15:00 on the second day. The petals began to close and wither before 13:00 on the third day (Fig. 1B).
Fig. 1.
Corolla opening size, nectar volume and nectar sugar concentration throughout the lifetime of a flower in the five oilseed rape cultivars. A, C & E: flower buds that were bagged at 18:00 and that opened before 9:00 the next morning, B, D & F: flower buds that were bagged at 9:00 and that opened at 11:00.
Data are presented as the means±standard errors.
Fluctuations in the nectar volume and the nectar sugar concentration were observed over the course of flowering (Fig. 1C–1F). For the flowers that were bagged at 18:00, there was a significant interaction between cultivar and time of day for nectar volume (Wald χ2=192.568, df=49, p<0.001) and nectar sugar concentration (Wald χ2=1731.946, df=45, p<0.001). The nectar volume showed a decreasing trend (Fig. 1C) and peaked at 9:00 (‘Ningde 21’: 1.133±0.351 μL; ‘Hengheyou 998’: 1.062±0.334 μL) or 13:00 (‘Aiyouwang’: 0.709±0.112 μL) on the first flowering day or at 9:00 (‘Dehuiyou’: 0.997±0.206 μL; ‘Huayou’: 0.930±0.272 μL) on the second day. The flowers wilted, and the nectar volume was lowest at 17:00 on the second day (Fig. 1C). The nectar sugar concentration increased between 9:00 and 13:00 on the first day and then decreased to a much lower value on the second day (Fig. 1E). The highest value (49.9%–54.1%) was observed between 13:00 and 15:00 on the first day. The lowest value was detected at 9:00 on the first day for ‘Aiyouwang’ (4.3±0.9%) and ‘Huayou’ (5.3±1.5%) and between 9:00 and 15:00 on the second flowering day for the other three cultivars (Fig. 1E).
Also for the flowers that opened at 11:00, significant interactions between cultivar and time of day were found for the nectar secretion volume (Wald χ2=853.836, df=59, p<0.001) and the nectar sugar concentration (Wald χ2=1229.683, df=59, p<0.001). Nectar volume on the second day was highest at 11:00 for ‘Hengheyou 998’ (1.491±0.161 μL) or at 15:00 for the other four cultivars (1.422±0.179–2.166±0.175 μL) (Fig. 1D).
The lowest nectar volume was detected at 11:00 on the first day for ‘Dehuiyou’ and ‘Huayou’ (0.120±0.051 μL and 0.127±0.032 μL, respectively). For the other three cultivars, the lowest values were found when the flowers were about to close and wilt during the period from 11:00–13:00 on the third day, range: 0.126±0.051–0.147±0.066 μL (Fig. 1D). The nectar sugar concentration showed a decreasing trend on the first day. Then, on the second day, the nectar sugar concentration first increased at the 9:00 and 15:00 time points and then decreased. On the third day, the nectar sugar concentration slightly increased. The nectar sugar concentration peaked (38.5±5.5%–41.5±5.8%) between 11:00 and 15:00 on the first day and reached the minimum value at 9:00 on the second day for ‘Dehuiyou’ (4.6±0.3%) and ‘Huayou’ (3.1±0.2%) or on the third day for the other three cultivars (Fig. 1F).
The flowers that were bagged at 18:00 secreted nectar for two days (Fig. 1C, Fig. 2A & 2C). There were significant interactions between cultivar and flowering day for daily nectar volume (Wald χ2=44.42, df=9, p<0.001) and daily nectar sugar concentration (Wald χ2=1060.971, df=9, p<0.001). The daily nectar volume and the daily nectar sugar concentration of ‘Aiyouwang’, ‘Ningde 21’, ‘Hengheyou 998’ and ‘Huayou’ were significantly higher on the first day (p<0.05). On both the first and second days, no significant difference was found in the daily nectar volume among cultivars (p>0.05) (Fig. 2A). On the first day, the daily nectar sugar concentration of ‘Aiyouwang’ (31.3±1.3%) was significantly lower than that of ‘Ningde 21’ (37.0±1.6%), ‘Hengheyou 998’ (38.6±2.6%) and ‘Huayou’ (38.4±1.9%) (p<0.05) (Fig. 2C).
Fig. 2.
Daily nectar volume and daily nectar sugar concentration in the five oilseed rape cultivars on different flowering days. A & C: flower buds that were bagged at 18:00 and opened before 9:00 the next morning, B & D: flower buds that were bagged at 9:00 and opened at 11:00. Different lowercase letters indicate significant differences among cultivars on the same flowering day (Kruskal-Wallis test or one-way ANOVA, p<0.05). Different capital letters indicate significant differences among different flowering days for the same cultivar (Mann-Whitney U test, independent samples t test or Kruskal-Wallis test; p<0.05).
The flowers that opened at 11:00 secreted nectar for three days (Fig. 1D, Fig. 2B & 2D). Significant interactions between cultivar and flowering day for daily nectar volume (Wald χ2=958.162, df=14, p<0.001) and daily nectar sugar concentration (Wald χ2=411.229, df=14, p<0.001) were detected. The daily nectar secretion volume was significantly higher on the second day for all cultivars (p<0.05) (Fig. 2B). On every flowering day, no significant differences were detected in the daily nectar volume or the daily nectar sugar concentration among cultivars (p>0.05). The daily nectar sugar concentration was significantly higher on the first day for all five cultivars (p<0.05) (Fig. 2D).
There was a significant interaction between cultivar and flower opening time (Wald χ2=172.346, df=9, p<0.001) for the total nectar yield. For all five cultivars, the total nectar yield of flowers opening before 9:00 (4.422±0.716–5.265±0.804 μL) was significantly lower than that of flowers opening at 11:00 (7.982±0.580–10.646±0.434 μL) (p<0.01) (Fig. 3A). For the flowers opening before 9:00, no significant differences were detected in the total nectar yield among cultivars (Kruskal-Wallis test: H=2.318, df=4, p=0.678). For the flowers opening at 11:00, the total nectar yield of ‘Huayou’ (7.982±0.580 μL) was significantly lower than that of ‘Ningde 21’ (10.646±0.434 μL) and ‘Dehuiyou’ (10.555±0.665 μL) (p<0.01) (Fig. 3A).
Fig. 3.
Total nectar yield and average nectar sugar concentration in the five oilseed rape cultivars. Group A: flower buds that were bagged at 18:00 and opened before 9:00 the next morning, Group B: flower buds that were bagged at 9:00 and opened at 11:00. Different lowercase letters indicate significant differences among cultivars for the flowers opening at the same time (Kruskal-Wallis test or one-way ANOVA, p<0.05). Different capital letters indicate significant differences between flowers of the same cultivar that opened at different times (Mann-Whitney U test or independent samples t test; p<0.05).
For the average nectar sugar concentration, a significant interaction between cultivar and flower opening time (Wald χ2=91.775, df=9, p<0.001) was detected. Among the cultivars, no significant differences in the average nectar sugar concentration were detected in the flowers that neither opened before 9:00 (F4, 65=2.187, p=0.080) nor opened at 11:00 (F4, 90=1.447, p=0.225). However, for ‘Ningde 21’, ‘Dehuiyou’, ‘Hengheyou 998’ and ‘Huayou’, the average nectar sugar concentration of flowers opening before 9:00 was significantly higher than that of flowers opening at 11:00 (p<0.01) (Fig. 3B).
When the flowers were in bloom, the corolla opening size most frequently ranged between 21 and 27 mm. The nectar volume (r: 0.296–0.436; p<0.001) and the nectar sugar concentration (r: 0.193–0.381; p<0.001) were positively correlated with the corolla opening size in all five cultivars (Fig. 4).
Fig. 4.
The relationships of nectar volume (A) and nectar sugar concentration (B) with corolla opening size.
For all five cultivars, none of the samples contained detectable amounts of sucrose. Nectar carbohydrates consisted almost exclusively of glucose and fructose. There were no significant differences in the contents of fructose (Kruskal-Wallis test: H=2.981, df=4, p=0.561) and glucose (F4, 25=1.505, p=0.231) among the five oilseed rape cultivars (Tab. 2). The nectar of ‘Dehuiyou’ contained less glucose (12.69±2.26 g/100 mL) than the nectar of the other cultivars. The quantities of glucose slightly exceeded those of fructose in ‘Ningde 21’ and ‘Hengheyou 998’. The average G/F ratio ranged from 0.89±0.09 to 1.44±0.38, and no significant differences were detected in the G/F ratios among cultivars (F4, 25=0.902, p=0.478) (Tab. 2).
The nectar sugar composition (g/100 mL) in the five oilseed rape cultivars
| 17.22±2.53 | 20.08±3.03 | 0.89±0.09 | |
| 20.30±2.81 | 18.40±2.84 | 1.30±0.29 | |
| 12.69±2.26 | 15.20±2.81 | 0.96±0.17 | |
| 17.56±1.31 | 16.52±3.69 | 1.44±0.38 | |
| 14.66±2.66 | 15.11±2.81 | 1.04±0.19 |
Within the air temperature range of 8.6–26.4°C and relative humidity range of 38.3–97.7%, the nectar sugar concentration was significantly positively correlated with air temperature and negatively correlated with relative humidity for all five cultivars (adjusted R2: 0.433–0.571; p<0.05). For ‘Dehuiyou’ (adjusted R2=0.025) and ‘Hengheyou 998’ (adjusted R2=0.027), the nectar volume had a significant positive correlation with air temperature and relative humidity (p<0.05). While the nectar volume of ‘Ningde 21’ had no significant correlation with air temperature and relative humidity (adjusted R2=0.010), the nectar volume of ‘Aiyouwang’ (adjusted R2=0.016) and ‘Huayou’ (adjusted R2=0.017) was significantly positively correlated with relative humidity (p<0.01) (Tab. 3).
Linear regression of environmental parameters correlated with oilseed rape flower nectar production
| % Relative humidity | 0.007 | 0.002 | 2.908 | 0.004 | 0.016 | ||
| % Relative humidity | 0.014 | 0.004 | 3.753 | <0.001 | 0.025 | ||
| Air temperature | 0.036 | 0.013 | 2.774 | 0.006 | |||
| % Relative humidity | 0.012 | 0.003 | 3.762 | <0.001 | 0.027 | ||
| Air temperature | 0.025 | 0.011 | 2.235 | 0.026 | |||
| % Relative humidity | 0.007 | 0.002 | 2.993 | 0.003 | 0.017 | ||
| % Relative humidity | −0.491 | 0.069 | −7.113 | <0.001 | 0.433 | ||
| Air temperature | 1.229 | 0.229 | 5.380 | <0.001 | |||
| % Relative humidity | −0.517 | 0.062 | −8.364 | <0.001 | 0.571 | ||
| Air temperature | 1.476 | 0.216 | 6.846 | <0.001 | |||
| % Relative humidity | −0.742 | 0.067 | −11.132 | <0.001 | 0.496 | ||
| Air temperature | 0.462 | 0.221 | 2.084 | 0.038 | |||
| % Relative humidity | −0.633 | 0.068 | −9.363 | <0.001 | 0.536 | ||
| Air temperature | 1.226 | 0.230 | 5.327 | <0.001 | |||
| % Relative humidity | −0.670 | 0.069 | −9.739 | <0.001 | 0.524 | ||
| Air temperature | 0.948 | 0.232 | 4.084 | <0.001 | |||
Note: In the table above, only the cultivars whose nectar volume and nectar sugar concentration were significantly correlated with air temperature or relative humidity are listed.
Nectar secretion patterns are regarded as an adaptive feature of insect-pollinated flowers and are an important aspect of the reproductive biology of plants (Pyke, 2016). Secretion of more nectar helps entomophilous plants to become more attractive to pollinators (Shu et al., 2019). Previous studies have examined the nectary size and the nectar carbohydrate production and composition of
Farkas (2008) studied nectar production and sugar composition in three oilseed rape cultivars in Hungary and reported that in some cases, nectar was found in blossoms between the bud and young flower growth stages. Chabert et al. (2018) also reported that the secretion of nectar in oilseed rape flowers started at the bud stage in a male-fertile hybrid F1 line and in its male-sterile parental line. Although the flower bud samples were limited, it could be confirmed in the present study that nectar secretion in oilseed rape flowers started at the bud stage. With a certain amount of nectar already available to attract pollinators, the onset of nectar secretion at the bud stage should perhaps promote cross-pollination at flower anthesis (Chabert et al., 2018). However, we still do not know precisely the cellular and molecular mechanisms that trigger the beginning and end of nectar secretion. The inconsistency in nectar secretion status may be due to the differences among individual plants or to different positions of the buds as has been shown for the variation in nectar production in
The nectar volume and nectar sugar concentration produced by
The investigation was carried out during the daytime, and the potential nectar yield that could be used by pollinators secreted by the flowers opening at different times was compared. For the flowers that were bagged at 18:00 and that opened before 9:00 the next morning, nectar secretion lasted two days, with significantly more nectar secreted on the first day. For the flowers that opened at 11:00, nectar secretion lasted three days, and significantly more nectar was secreted on the second day. This is consistent with the finding that young and pollen-shedding flowers are the best nectar producers (Farkas, 2008). Furthermore, it has been verified that the removal of nectar could promote the secretion of more nectar (Torres & Galetto, 1998), and that when the nectar in the corolla reaches its maximum capacity and is not utilized, it will then be reabsorbed (Burquez & Corbet, 1991; Nepi & Stpiczyńska, 2008). In our study, the total daytime nectar secretion volume of flowers opening before 9:00 was significantly less (p<0.01) than that of flowers opening at 11:00. This might be due to the longer flowering period of flowers opening at 11:00, which allows for more nectar secretion and nectar collection. Our result that the average nectar sugar concentration of flowers opening before 9:00 was higher than that of flowers opening at approximately 11:00 was consistent with the finding of a previous study by Mohr & Jay (1990) that more rounds of extraction increased the amount of nectar but decreased the sugar concentration of nectar.
Chabert et al. (2018) have been reported that nectar secretion is synchronized with stigma receptiveness. In this study, we detected a significant positive correlation between corolla opening size and both nectar volume and the sugar concentration of nectar. However, the relevance of corolla opening size and stigma receptiveness has yet to be established. It is important to note that our study was conducted only during the daytime. The study period did not cover the whole flower lifetime for the flowers opening before 9:00. Further studies are required to determine the period of greatest pistil receptivity and to identify the relationship of the pistil with flower characteristics throughout a flower’s lifetime, taking flower longevity variation into account.
Oilseed rape nectar is mainly composed of glucose and fructose (Kevan et al., 1991; Farkas, 2008). In this study, we detected that the nectar carbohydrates of the five cultivars consisted almost exclusively of glucose and fructose. Importantly, honey crystallizes quickly if the G/F ratio is more than 0.90 (Smanalieva & Senge, 2009). Oilseed rape honey tends to granulate readily when the G/F ratio reaches or exceeds 0.90 in most cultivars (Kevan et al., 1991; Farkas, 2008; the present study). With large production and the tendency to granulate quickly, the price of oilseed rape honey is much cheaper than that of the other kinds of honey (Gao & Zhao, 2018). However, not all oilseed rape crops produce honey of equal granulating tendency, as sugar ratios of different varieties vary (Kevan et al., 1991). Information on the sugar composition of different varieties can help breeders to select oilseed rape cultivars that produce honey with a low tendency to granulate.
Microclimatic conditions have been found to influence nectar production (Walker et al., 1974; Pinzauti, 1986; Petanidou & Smets, 1996). When the external temperature has not yet reached the optimal temperature for nectar secretion, a proper increase in temperature can promote secretion (Takkis et al., 2018). Nectar production has been found to be dependent on ambient humidity, with nectar volume increasing with increasing humidity (Bertsch, 1983). In our study, we found that the nectar secretion of oilseed rape flowers is affected by the combination of air temperature and relative humidity. For two of the five cultivars, the nectar secretion volume was positively correlated with air temperature and relative humidity. Slight differential effects of environmental parameters on nectar secretion among cultivars were also detected in this study. Our results indicated that the nectar sugar concentration was positively correlated with air temperature and was negatively correlated with relative humidity. The consistent trend of the correlation between the nectar sugar concentration and the relative humidity has been demonstrated in previous studies on other plant genera (Corbet et al., 1979; Bertsch, 1983).
In the oilseed rape breeding process, higher nectar volumes and nectar sugar concentrations in flowers are more favourable for attracting insects for pollination, which in turn promotes an increase in oilseed rape yield (Bommarco et al., 2012; Shu et al., 2019). Entomophilous plants use nectar to attract pollinating insects, and nectar secretion patterns can be correlated with bee activity to analyse the interactions between pollinators and insect vector plants during pollination. Analysis of nectar yield in flowers can help predict potential honey production and inform strategic allocation of bee colonies.