The Effective Pollination Period of European Plum ( Prunus domestica L.) Cultivars in Western Norway

: This study evaluated the effective pollination period (EPP) in four European plum ( Prunus domestica L.) cultivars (‘Mallard’, ‘Edda’, ‘Jubileum’, and ‘Reeves’) during two years (2018–2019) under the environmental conditions in western Norway. The pollination of plum cultivars was carried out one, three, ﬁve, seven, and nine days after anthesis (DAA) with a pollen mix of two compatible cultivars (‘Victoria’ and ‘Opal’). Initial, middle-season, and ﬁnal fruit set was recorded after one month and two months after pollination and just before the harvest, respectively. On average from both years cultivar ‘Jubileum’ had the highest fruit set when pollinated one, three, ﬁve, seven, and nine DAA (33.23%, 30.83%, 8.47%, 3.08%, and 1.15%, respectively), which was more than two folds higher fruit set than in the other studied cultivars. Cultivar ‘Jubileum’ showed signiﬁcantly reduced fruit set between pollination on ﬁve and nine DAA, while cultivars ‘Mallard’, ‘Edda’, and ‘Reeves’ had markedly reduced fruit set if pollinated three to ﬁve DAA, implying that the EPP in ‘Jubileum’ was ﬁve days while in the rest it was three days. Variation of weather conditions during the ﬂowering period in both years did not have a major effect on the receptivity of stigmas in the studied plum cultivars, which means that the existing differences in the length of EPP is maternal-genotype dependent.


Introduction
There are two important species of plum in commercial horticulture, namely European plum and prunes (Prunus domestica L.) and the Asian or Japanese plums (Prunus salicina). Both belong to the Rosaceae family and have been cultivated for at least 2000-4000 years.
The European plum is a very important temperate zone fruit species, but can be grown both in a cooler climate and in sub-tropical conditions. The P. salicina plums are mostly grown in China and P. domestica in southern Europe. However, P. domestica is well adapted to cooler climate such as northern Europe. The world plum production is 2.7 million ha and, with 12.6 million t. China is the leading country (>7 million tons on >2.1 million ha) followed by Romania (>692,000 tons, >65,000 ha) and Serbia with >558,000 tons on >72,000 ha [1]. Regardless of the wide geographic distribution of plum cultivars over various climate zones in each area only specific cultivars can be grown [2]. Independent of countries, the plum production is mainly destined for fresh consumption, drying, preserves, and brandy making.
Commercial fruit production in Norway is located in the fjord districts of the southwestern part and around lakes at eastern part of the country at latitude around 60 • north. This represents the world's northern border of commercial fruit production. The climate, where the temperate fruits are grown, is closely linked with the Gulf Stream which flows The length of the EPP can vary due to the flower quality, flower position within the inflorescence, and also due to certain cultural practices such as chemical treatments [16]. In diploid cultivars, the egg apparatus is 'mature' and ready to be fertilized during flower opening, but in triploid cultivars the maturation is delayed until two to three days after anthesis, which results in extended ovule longevity and a longer EPP [17]. Besides, other abiotic factors, in most cases temperature, can speed up or slow down the span of EPP [16].
The profitability of plum growing is strongly affected by fruit set, and the pollination must happen in the right moment in order to have satisfactory yields. EPP has been proved to have great impact on post-fertilization fruit drop, and it plays a significant role in yearto-year cropping variability, which taken together influences yields. Thus, the aim of this study was to determine the effective pollination period in four plum cultivars grown in a Nordic climate.
The trees were planted in 2012 and all cultivars were grafted on the rootstock St. Julien A. The tree row orientation was South-North in a Southern slope; trees trained as slender spindle trees and limited 2.5 m height. Orchard floor management consisted of grass in the inter-rows and a 1-m wide vegetation-free strip in the intra-row spaces (wowed plastic). The trees were irrigated by drip irrigation when water deficits occurred.

Air Temperature and Precipitation
West Norway, especially the fjord areas have a cool, maritime, Nordic climate which is under the influence of the Gulf Stream. Summers are cool and winters are mild. Leikanger, where the experiment was conducted, is located in the county of Vestland, along the Sognefjord, West Norway. Average annual air temperature is 7.3 • C (during spring 4-11 • C), annual rainfall is 1063 mm (with May being driest with 15 rainy days) and an average annual humidity percentage of 78.0%. Average monthly hours of sunshine are the highest in May (~175 h) and the mean monthly wind speed over the year is 3 m/s. Unfavorable environmental conditions, especially cold temperatures, fog and rain can often occur during spring. For the purpose of this study daily temperatures (average, max and min) and precipitation (mm) were recorded during flowering period of plum cultivars (https://lmt.nibio.no/station/35/, accessed 20 December 2021) and presented on

Flowering, Pollen Collection and Pollen Germination In Vitro
Flowering stages for studied plum cultivars were recorded according to the BBCH scale [19]. The beginning of flowering (BBCH stage 61) was the date when 10% of flowers were open, full bloom (BBCH stage 65) was determined when 50% of flowers were open, while end of flowering was noticed when the majority of petals were fallen (BBCH stage 67). Determination of flowering in all plum cultivars studied herein was carried out on five trees per cultivar.
Pollen was gathered in two consecutive years (2018 and 2019). A mixture of the freshly gathered pollen of the two plum cultivars ('Victoria' and 'Opal'), being fully compatible as proved by Meland et al. [20], were used as pollinizers in both years and for all four tested cultivars. Pollen sampling and in vitro pollen germination was carried out according to Fotirić Akšić et al. [21]. Pollen gathering was carried out from four trees per pollinizer by collecting flowers from all around the canopy.

Emasculation and Pollination
In order to determine effective pollination period, the pollination was carried out one, three, five, seven, and nine days after anthesis (DAA). Tents for isolation were placed over the trees in the 'balloon' phase (BBCH stage 59). Emasculation for studying effective pollination period were carried out in cultivars 'Edda', 'Mallard', 'Reeves', and 'Jubileum'. Flowers in the balloon stage were emasculated by removing the whole perianth. About 20 flowers per 5 replicates were emasculated per each cultivar and per each term of pollination. Every branch was labeled with date of emasculation, replication and planned pollination day. The hand pollination of emasculated flowers was carried out first by dipping a finger into the mixture of two-plum-cultivars pollen and then by touching the exposed stigma two times. Pollination was considered successfully carried out when yellowness was observed on the stigma. At pollination, the number of pollinated flowers was counted, and the number of pollinated flowers was written on the labels. If an open, un-emasculated flower appeared, it was eliminated immediately. The tents were removed one week after the last pollination.

Fruit Set
The percentage of initial (IFS), middle-season (MFS), and final fruit set (FFS) was counted one month, two months after pollination and just before the harvest, respectively, in both years. Fruit set (IFS, MFS and FFS) was calculated as follows: IFS/MFS/FFS = (number of developing fruitlets/total number of flowers) × 100.

Statistical Analysis
The data obtained for fruit set were statistically analyzed using Fisher's model of twofactor analyses of variance (ANOVA). The significances of the individual differences for the investigated factors were determined using the least significant difference (LSD 0.05 = 95% confidence). Statistical analyses were conducted using STATISTICA for Windows 6.0 (StatSoft Inc., Tulsa, OK, USA).

Meteorological Data
The weather conditions during the two-year experiment differed significantly. Average maximal and minimal temperatures in 2018 were 24.8 • C and 13 • C, respectively, while in 2019 it was 10.3 • C and 3.2 • C. The average vegetation temperature (from April to September) was 13.6 • C in 2018 in contrast to 12.9 • C in 2019. Much larger discrepancy between ecological factors between two years was noticed in the 10-days after flowering period. In this period in 2018, the mean average daily temperatures were 18.5 • C compared to 2019, where average daily temperatures were 6.5 • C (Figures 1 and 2).
Regarding precipitation, in 2018 ( Figure 1), only one spring shower occurred during pollination (7.8 mm/day) with no rain ten days afterwards. In 2019 (Figure 2), three short rainfalls took place during the pollination (1.5 mm/day; 0.6 mm/day and 1.6 mm/day) and three in a 10-days-after-pollination period (0.7 mm/day; 7.1 mm/day and 0.4 mm/day).

Flowering and In Vitro Pollen Germination
In 2018, all plum cultivars were flowering in May (Figure 3), while in 2019 it was two weeks earlier (in April). The sequence of the cultivar's flowering was the same in both years. A period from full flowering (BBCH stage 65) till petal fall (BBCH stage 67) for all studied cultivars and both years lasted six to seven days [22]. between ecological factors between two years was noticed in the 10-days after flowering period. In this period in 2018, the mean average daily temperatures were 18.5 °C compared to 2019, where average daily temperatures were 6.5 °C (Figures 1 and 2). Regarding precipitation, in 2018 ( Figure 1), only one spring shower occurred during pollination (7.8 mm/day) with no rain ten days afterwards. In 2019 (Figure 2), three short rainfalls took place during the pollination (1.5 mm/day; 0.6 mm/day and 1.6 mm/day) and three in a 10-days-after-pollination period (0.7 mm/day; 7.1 mm/day and 0.4 mm/day).

Flowering and In Vitro Pollen Germination
In 2018, all plum cultivars were flowering in May (Figure 3), while in 2019 it was two weeks earlier (in April). The sequence of the cultivar's flowering was the same in both years. A period from full flowering (BBCH stage 65) till petal fall (BBCH stage 67) for all studied cultivars and both years lasted six to seven days [22]. In both years, cultivars 'Victoria' and 'Opal', whose pollen was mixed and used for testing effective pollination period of 'Mallard', 'Jubileum', 'Edda' and 'Reeves', flowered in the same interval. Pollen germination rates of cultivar 'Victoria' in both years were very high (91.2% in 2018, and 72.3% in 2019), while cultivar 'Opal' had a bit lower pollen germination rate in 2018 (22.4%), but a quite high in 2019 (54.1%) [22].
In the first year of the study, cultivars 'Mallard', 'Edda', and 'Reeves' presented the maximum of IFS, MFS, and FFS on the one DAA (Figure 4). For the cultivar 'Jubileum' the maximum of IFS, MFS, and FFS was on the three DAA, and significantly higher from the results obtained for fruit set on the one DAA. In 2019, behavior of the plum cultivars 'Mallard' and 'Reeves' was similar. The situation with 'Edda' was a bit different, which maximum IFS and MFS were three DAA, while FFS was on the one DAA. Also, 'Jubileum' changed, and its IFS, MFS, and FFS was the highest on the 1 DAA.    The evaluation of the EPP in 2018 using sequential pollination showed that plum cultivars 'Mallard', 'Edda', and 'Reeves' achieved maximum FFS when pollinated in 1 DAA (10%, 8%, and 6.4%, respectively), and significantly lower but still satisfactory FFS (>5%) in 3 DAA (6.2%, 5.6%, and 5%, respectively). The cultivar 'Jubileum' had the highest FFS in three DAA (39.93%), while in one DAA (37.80%) and five DAA (11.6%), the results were significantly lower but still considerable valuable for a plum cultivar. In the second year, all studied cultivars had the highest FFS in one DAA, but satisfactory fruit set in three DAA, while only cultivar 'Jubileum' had ably fruit set >5% even in five DAA (5.33%). From all of this can be estimated that the EPP for 'Mallard', 'Edda', and 'Reeves' is three days, while in 'Jubileum', which gives acceptable yields in five DAA, is five days ( Figure 5). The evaluation of the EPP in 2018 using sequential pollination showed that plum cultivars 'Mallard', 'Edda', and 'Reeves' achieved maximum FFS when pollinated in 1 DAA (10%, 8%, and 6.4%, respectively), and significantly lower but still satisfactory FFS (>5%) in 3 DAA (6.2%, 5.6%, and 5%, respectively). The cultivar 'Jubileum' had the highest FFS in three DAA (39.93%), while in one DAA (37.80%) and five DAA (11.6%), the results were significantly lower but still considerable valuable for a plum cultivar. In the second year, all studied cultivars had the highest FFS in one DAA, but satisfactory fruit set in three DAA, while only cultivar 'Jubileum' had ably fruit set >5% even in five DAA (5.33%). From all of this can be estimated that the EPP for 'Mallard', 'Edda', and 'Reeves' is three days, while in 'Jubileum', which gives acceptable yields in five DAA, is five days ( Figure  5).

Flowering
Awareness of flowering time of plum cultivars is fundamental for a proper choice of cultivar combination that provides successful pollination and fertilization for acceptable fruit set [11]. Thus, in case that plum cultivar(s) bloom too early with short flowering period, and without overlapping with other cultivars, those should be avoided due to the poor fruit set [23]. The length of the flowering period in plums is both genetically inherited and a result of the influence of the climatic conditions. Most European plum genotypes

Flowering
Awareness of flowering time of plum cultivars is fundamental for a proper choice of cultivar combination that provides successful pollination and fertilization for acceptable fruit set [11]. Thus, in case that plum cultivar(s) bloom too early with short flowering period, and without overlapping with other cultivars, those should be avoided due to the poor fruit set [23]. The length of the flowering period in plums is both genetically inherited and a result of the influence of the climatic conditions. Most European plum genotypes have simultaneous flowering periods, providing an adequate overlap for effective cross-pollination among them [24]. Corresponding to Koskela et al. [11], overlapping is considered too short if the difference between the dates of full bloom is six days or more. The duration of flowering in the studied plum cultivars was seven to nine days in 2018 and seven to eight days in 2019, which is in accordance to Szabo and Nyeki [6] who stated that the length of blooming in Prunus domestica is from six to eleven days. According to those authors, temperatures before flowering are not as important as those after anthesis, so there is negative correlation between the mean temperatures and the blooming period. The overlapping of flowering in the studied cultivars with pollen donors ('Victoria' and 'Opal') in the climatic conditions of western Norway has been already proved by Meland et al. [20] and Cerović et al. [22].

Effective Pollination Period
In fruit trees, the EPP varies a lot and depends on species and cultivars, nutritive state of the tree, crop load, chemical treatments and alternate bearing [12]. The duration of the EPP is influenced by many factors which include pollen tube growth, stigmatic receptivity and ovule longevity period. Low fruit set and EPP in some Prunus species has been related to unfavorable environmental conditions of the localities and seasons such as low and or high temperatures before and after flowering, winds, and drought [16,[25][26][27]. The time required for pollen tubes to grow from stigma to the base of styles, and later enter to the micropila and carry out fertilization, decreases with increasing day/night temperature from 13/2 to 24/7 • C [28]. At excess temperatures, the number of pollen tubes that reach the base of the style can decline [29]. Also, manipulation with the flowers (mostly emasculation) can shorten EPP since damage the flowers can cause flower and/or fruitlet drops due to premature degeneration of ovules [30][31][32].
The EPP in 'Mallard', 'Edda', and 'Reeves' lasted for three days, whereas, in 'Jubileum' was longer and lasted five days. No matter that the environmental conditions during sequential pollination of plum cultivars were different, the life span of egg cells in the studied four plum cultivars were constant and maternal-genotype dependent. Similar findings were found by Guerra et al. [48] for some Japanese plum cultivars, which had a stable response to factors that caused ovule degeneration. Our results coincide with Keulemans and van Laer [49] who proved that EPP in plum cultivars 'Monsieur Hatif' and 'Bleue de Belgique', with thresholds of 10% fruit set, was six and three days, respectively.

Fruit Set
Fruit set depends on many factors (climate, source of pollinators, pollination, pollen tube growth, ovule longevity, compatibility between the cultivars, overlapping of the flowering periods, and others) but pollen germination is one of the main prerequisites to have high yields [50]. Since fruit trees of the Rosaceae depend on cross-pollination and, therefore, commercial orchards need to contain at least two cross-compatible cultivars with synchronous flowering, we picked two compatible cultivars as pollinizers [51]. Both cultivars that were used as-pollinizers showed satisfactory pollen germination which is consistent with the values obtained for the pollen germination in vitro for numerous plum cultivars that ranged from 12.96 to 71.6% [52][53][54].
Kron and Husband [55] reported that an increased number of pollen donors might promote seed set and reduce seed abortion. Since most research studies evaluated a single pollen donor at a time, a pollen mixture of two compatible cultivars was applied for EPP determination in this study. Besides that, a long time ago Keulemans and Van Lear [49] claimed that cultivars 'Opal' and 'Victoria' show good pollen germination and fast growth of pollen tubes (even at low temperatures).
For the family Rosaceae, Janssen et al. (2008) [56] and Bonghi et al. (2011) [57] proved that reproductive biology (flower buds' production and decline, flowering time, synchrony between pollen tube arrival to the ovule and embryo sac maturation, syngamy, and successful early embryo development) and fruit set are determined by genes however, external environmental factors play important role too. In numerous fruit species and/or cultivars, stigma is receptive during anthesis. Unfortunately, however, the life span of the receptivity differs between species, and if it is too short it can jeopardize the fruit set [47]. Besides, abiotic factors (temperature, rainfall, wind, and drought) and nutritive stage of the flower can influence fruit set too [22,58].
In both years of the experiment values of IFS were much higher than the values of MFS and FFS (Figure 4), which implies that the physiological fruit drop happened before counting MFS. In 2018 this happened before 10 July and in 2019 before 25th June. In all studied plum cultivars, FFS was~five-folds lower than IFS. This is expected since FFS is affected by the management factors such as crop loads, nutrients, irrigation, type of training system, and many more. Similar results were obtained by Ðordević et al. [24] for 'Pozna Plava', 'Čačanska najbolja' and 'Presenta', where FFS compared to IFS was two-to five-folds lower.
Comparing the average of IFS in both years, it can be concluded that the cultivar 'Jubileum' had the highest fruit set both in all terms of pollination, and averagely for all treatments (2.2 fold higher than 'Edda', 1.8 folds higher than 'Mallard' and~2.75 folds higher than 'Reeves'). High fruitfulness of this cultivar that is widely grown in Norway was already proved by Meland and Birken [59], who claimed that it initiates too many flowers and set too many fruits. The highest fruit set was determined on one DAA ('Edda', 'Mallard' and 'Reeves') and three DAA ('Jubileum') with a progressive decrease thereafter. It is very important that the stigmas are receptive at or shortly after flower opening since the first flowers are inclined to give fruits of the best size and quality [11].
By calculating just FFS, averagely for both years, cultivar 'Jubileum' gave the highest values of fruit set in all treatments, 13.3% fruit set when pollinated in one DAA, 11.6% in three DAA, 7.25% in five DAA, 1.2% in seven DAA and 0.7% in nine DAA. This goes in the line with the findings of Sanzol and Herrero [16] who found that in some fruit species/cultivars, good correlations have been found between the percentages of fruit set following delayed pollination and ovule development, pollen tube growth, and stigma receptivity. The rest of the studied cultivars had a maximum fruit set of one DAA. The very low fruit set in seven DAA (0.5 to 3.08%) and in nine DAA (0 to 0.15%) implies that either the egg cell was not receptive or the style and transmitting tissue inside was not able to support pollen tube growth.
The discrepancy in the fruit set between the cultivars, besides stemming from a genetic component, could be due to the temperature differences during the blooming time and in the period after pollination. In 2018, 'Mallard' bloomed four days before 'Reeves' and, in 2019, it bloomed two days earlier. In both gap-periods, temperature maximums were over 20 • C. Generally, high temperatures can alter the soluble carbohydrate content in the pistil, which decreases the pollen tube growth along the style or ovule fertility and reduces in fruit set [60]. In post pollination periods in 2018, temperatures were over 20 • C, while in 2019 the daily maximum did not reach 10 • C. According to Weinbaum [61] temperatures below 10 • C can cause a loosening of the contact between the pollen grain and tube membrane.

Correlation between Effective Pollination Period and Fruit Set
To evaluate the correlation between fruit set and EPP, the pattern of average fruit set on sequential pollination was plotted against the pattern of EPP. EPP was positively and significantly correlated with each pollination time but the highest coefficient of correlation was between EPP and one DAA (r = 0.83, p = 0.011). This correlation also decreased with the delay in pollination time. These results indicated that the length of EPP depend the most on fruit set immediately after flower opening (with 'Jubileum' as an exception where delayed pollination gave acceptable results too). The reason of such a short EPP in 'Mallard', 'Edda', and 'Reeves' could be due to the degeneration and low viability of the reproductive organs. A tendency towards early ovule degeneration, both in pre-fertilized and post-fertilized stages, may negatively affect fertilization and fruit set. A short life span of plum ovules has been already reported in plums [17]. Egea and Burgos [62] and Kodad and Socias i Company [63] defined a term 'style receptivity period' or 'style suitability period' which can also limit EPP. In case of delayed pollination, maybe egg cells in the studied plum cultivars were still viable but stigma senescence prevented pollen germination and fertilization of the ovule. Short EPP that was provoked by early stigma degradation was previously determined in sweet cherry [64], apricot [65], pear [35], and almond [63].
Unfavorable weather conditions at blooming time are known to have a detrimental effect on fruit set by reducing the germination of pollen on the stigma and its growth down the style into the ovary and minimize insect activity. Prolonged EPP of the 'Jubileum' is extremely important for the west Norwegian and/or similar climatic condition, especially in the changing weather during springs (with often showers, north winds, high humidity, and low temperatures), when fertilization and fruit set can be endangered. Under this kind of weather five-days-long EPP of 'Jubileum' can give an opportunity to this cultivar to 'wait for better days' and set acceptable fruit set and secure satisfactory yield.

Conclusions
Stigma receptivity in studied plum cultivars was adequate during anthesis, and plum flowers were able to produce optimum fruit set when pollinated at the beginning of flowering. As the flowers were aging, the fruit set was markedly reduced, especially from three to five DAA, in 'Mallard', 'Edda', and 'Reeves' and from five to seven DAA, in 'Jubileum', so it can be concluded that EPP in these cultivars was three and five days, respectively. Since this study was conducted in the west of Norway, the results can be used for different locations with similar climatic conditions. These finding have practical implications in plum production. It is relevant to know until which day the pollination has to be carried out in certain cultivars in order to secure satisfactory yields. Regarding future breeding work, only cultivars (in this case 'Jubileum') that show long EPP should be used as parents for creating new plum cultivars to obtain genotype which will have high ovule longevity and/or stigma receptivity. Only this kind of genotypes can ensure regular and high yields (especially in the changing environmental climate such as west Norway).