ORIGINAL ARTICLES
Effects of Pollination of Some Stigmas in Kiwifruit Flowers on Seed Distribution and Fruit Quality
2025 Volume 94 Issue 2 Pages 184-189
Details
2025 Volume 94 Issue 2 Pages 184-189
Kiwifruit flowers have approximately 40 stigmas. Artificial pollination is essential for dioecious kiwifruit production. We investigated the effect of the pollination site on seed distribution, fruit size and quality. We found that seeds should be evenly distributed in the fruit to improve fruit shape and that seed number and total soluble solids (TSS) were not positively correlated. We also noted that seeds were distributed in a spreading manner away from pollinated stigma. This indicated that the stigma and carpel had a similar one-to-one like connection. However, the connection was not exact, and seeds were distributed not only in a particular carpel, but also slightly the outside the carpel. Seeds were distributed uniformly in the fruit even when pollination was limited to the two most distant stigmas, and the number of seeds increased when the number of pollinated stigmas increased. These results indicated that it was not necessary to pollinate all stigmas, but only several stigmas, in order to achieve larger fruit size.
Kiwifruit is a dioecious plant that does not self-pollinate. Therefore, pollination by flower-visiting insects or artificial pollination is necessary for commercial fruit production. Pollen for artificial pollination is collected manually in advance and the pollen collection process is susceptible to environmental factors. Imported pollen is also used for artificial pollination. However, securing imported pollen has become difficult owing to the global spread of bacterial diseases (Scortichini et al., 2012), and soaring pollen prices caused by the imposition of strict import quarantine measures to prevent the spread of diseases caused by bacteria-contaminated pollen. For these reasons, it has become difficult to secure pollen, and it is necessary to reduce the amount of pollen used.
Kiwifruit has numerous seeds within a single fruit and requires adequate pollination to ensure fruit set and produce high-quality fruit (Ohara et al., 1997). As fruit size is linked to market price (Pescie and Strik, 2004), there have been many studies on the relationship between the number of seeds and fruit size. Generally, more than 800 seeds are required for the fruit to grow to more than 100 g in weight (Costa et al., 1993). To develop a simple technique for artificial pollination, various types of intensifiers and methods to reduce pollen quantity have been studied together (Fujishima et al., 2006; Matsumoto et al., 2007; Murakami and Yamaguchi, 2017; Waki et al., 2000). In Japan, some local governments have started pollen production. However, the amount produced is still insufficient.
Although the formation of deformed fruit is thought to be due to uneven seed distribution, no detailed investigations have been carried out. The pistil of a female kiwifruit flower has approximately 40 stigmas (Fig. 1A). Pollination is known to have an impact on seed distribution. However, no studies have been performed to clarify the effect of pollination of only some stigmas on fruit quality, specifically fruit size and the number of seeds. Although it is assumed that uneven pollination affects seed distribution, there are no reports on how it affects fruit size and the number of seeds. In order to develop new pollination methods and reduce the amount of pollen used, it is necessary to clarify the mechanism underlying the process from pollination to fruit development. In this study, we aimed to clarify the relationships between pollination and seed distribution and between pollination and fruit growth.
Kiwifruit pistillate flower with partial stigmas. Stigma (arrowhead) and seared stigma (encircled). A: Untreated flower (control). B: “one-place treatment”, one or two stigmas at one place. C: “opposite-end treatment”, one stigma is on the opposite side of the other stigma.
‘Hayward’ kiwifruit (Actinidia deliciosa (A. Chev.) C. F. Liang & A. R. Ferguson) were planted in the Agricultural Farm of Meijo University located in Kasugai, Aichi Prefecture, Japan. ‘Tomuri’ was used as the pollinizer. Pistillate flowers were covered with paper bags before flowering to avoid natural pollination.
Partial pollinationThe following partial pollination treatments were performed. In the “one-place treatment”, one or two adjacent stigmas in one place were pollinated (n = 60 in 2014; n = 60 in 2017; n = 150 in 2018; and n = 15 in 2019). In the “half treatment”, all stigmas on one side were pollinated (n = 60 in 2014). In the “opposite-end treatment”, a stigma and the other stigma on the opposite side were pollinated (for a total of two stigmas on a flower; n = 7 in 2019). Stigmas other than pollinated ones were cut at half the length of the stigma with scissors, and the cut ends were seared with a soldering iron to inhibit pollination (Fig. 1B and C). Pollination was accomplished with a fine watercolor brush using dried pollen collected in advance. As a control, all stigmas of a normal pistillate flower were pollinated using previously collected staminate flowers. Fruit set for all treatments was confirmed in June. Pollen germination rates were 64.4% in 2014, 81.1% in 2017 and 48.2% in 2018. In the “non-pollination treatment”, flowers were covered with paper bags during the flowering period; these treatments were performed in 2014 (n = 60), 2017 (n = 10), and 2018 (n = 45). In the other treatment, the paper bags were temporarily removed for a few minutes for artificial pollination. In the no-stigma treatment, all stigmas were cut and soldered before artificial pollination in 2017 (n = 10). In the “control”, flowers were not covered with paper bags. To confirm whether fertilization of the seared stigmas was possible, all stigmas were seared and pollinated manually. All treated flowers were bagged until June when all staminate flowers dropped and physiological fruit drop was completed.
Fruit growthFruit set was determined by removing the bags in June when physiological fruit drop finished. Fruit longitudinal length and long width were measured approximately every seven days in 2014 (n = 12), 2017 (n = 20), and 2018 (n = 20). One fruit (2.2%) with seeds set in 2018. This may have been due to pollen contamination in the paper bag for some unknown reason.
Number of seeds in a fruitThe number of seeds in fruit derived from the treatments in 2014 (n = 5), 2017 (n = 5), 2018 (n = 5), and 2019 (n = 7) was investigated. The fruit was cut into five equal transverse sections (Fig. 2), and the number of seeds in each section was counted. In 2018, each transverse section was further divided into four compartments: a compartment with accelerated fruit enlargement near the equatorial plane, that with suppressed fruit enlargement, and the middle of those two compartments. The number of seeds in each compartment was determined vertically from the stylar end.
Kiwifruit was cut into five transverse sections, and each section was further divided into four compartments in 2018. A shows five equal transverse sections. B outlines the four compartments in each section.
Fruit quality was measured immediately after harvest and after ripening. Ripening was promoted with a commercial ripening agent sachet (ethylene). The juice was collected from the equatorial plane of each fruit. Total soluble solids (TSS) were determined by differential refractometry. Titratable acidity was determined by titration with 0.1 N NaOH. Titration values were converted into citric acid. The NaOH titer was determined by volumetric analysis using HCl before analysis of the juice.
Starch content was determined using the pulp after the juice was collected. Seeds were removed from the pulp, crushed, and immersed in 70% ethanol for 24 hours at 4°C to extract soluble substances. This process was repeated four times. The extraction residue was dried in a centrifugal decompressor. The starch content in the dried sample was determined using a Total Starch Assay Kit (Megazyme International Irelands Ltd., Bray, Ireland).
Fruit pulp color after ripening was determined by measuring L* and a* values on the equatorial plane using a colorimeter (ZE2000; Nippon Denshoku Industries, Co., Ltd., Tokyo, Japan).
A high fruit set ratio of 76.0% was noted in the opposite-end treatment in 2018 (Table 1). In contrast, all of the non-pollinated flowers dropped in 2014 and 2017. All of the no-stigma flowers also dropped in 2017.
Fruit set ratios of non-pollinated flowers and pollinated kiwifruit pistillate flowers with opposite-end stigmas or no stigma.
The growth curve of fruit longitudinal length was the highest for the control and the lowest for the one-place treatment in 2014 (Fig. 3). The same tendency was observed for the fruit growth curves in 2017 and 2018 (data not shown). Comparing fruit size at harvest, the longitudinal lengths of the control were always larger than those of partial pollination treatments. The long width of the control was significantly greater than in the partial pollination treatments, except in 2014 (Table 2).
Changes in longitudinal length of kiwifruit derived from one-place treatment of stigmas (▲), half treatment of stigmas (■), and a normal pollination (control, ●) in 2014. Vertical bars indicate SE (n = 12). Different letters indicate a significant difference (P < 0.05, Tukey-Kramer test).
Effect of pollination treatment on fruit longitudinal length and long width.
The fruit from the one-place treatment was small with dropped shoulders and had a small number of seeds on the dropped shoulder side (Fig. 4). In the half treatment, the fruit was straight and large, but the number of seeds on the equatorial plane was smaller than that of the control (Table 3). Fruit from the one-place treatment and control fruit were divided transversely into five equal sections to observe the distribution of seeds (Fig. 5). Seeds were unevenly distributed in all sections of fruit from the one-place treatment. The numbers of seeds in the enlarged and suppressed compartments of the equatorial plane, below equatorial plane, and the stylar end were significantly different (Table 4). There was a significant difference (P < 0.01, t-test) between the total number of seeds in whole fruit of the control and the one-place treatment. The number of seeds in the control and one-place treatment fruits were 979.2 and 630.5, respectively. The total number of seeds in each section was smaller than that in the control. The control fruit had many seeds in all sections, and the seeds were distributed evenly.
Appearance of kiwifruit and their equatorial planes, derived from different treatments in 2014. A and a: One-place treatment. B and b: Half treatment. C and c: All stigmas pollinated (control).
Number of seeds in kiwifruit subjected to pollination treatment in 2014, 2017, 2018 and 2019.
A kiwifruit was cut into five transverse sections in 2018. The rightmost and leftmost sections are the peduncle and stylar end sections, respectively. A: One-place treatment. B: All stigmas pollinated (control).
Number of seeds in each compartment of various transverse sections of kiwifruit derived from one-place treatment and the control (normal pollination) in 2018.
No differences in TSS, acidity, starch content, or flesh color (L* and a*) were noted between one-place treatment fruit and control fruit before and after ripening (Table 5).
Fruit quality before and after ripening, indicated by TSS (total soluble solids), acidity (citric acid, titratable acidity), starch, and flesh color (L*, a*) for two compartments on the equatorial plane, for one-place treatment and normal pollination (control) in 2018.
It is known that seed number is important for kiwifruit enlargement and that pollination is essential for securing seeds (Costa et al., 1993; Murakami and Yamaguchi, 2017). Pollination is completed when pollen is deposited on a stigma. However, it is not clear how pollination of its 40 separate stigmas affects kiwifruit size. Therefore, we investigated the effects of artificial pollination on fruit growth and quality by restricting pollination to some selected stigmas.
The kiwifruit flower has approximately 40 stigmas, and the estimated number of carpels in fruit ranges from 26 to 38 (Hopping, 1976). Each carpel is connected to a stigma, which is a multicarpellary organ consisting of an unspecified number of carpels, instead of a one-to-one pair (Howpage et al., 1998). For one-place treatment in which one or two stigmas in one place were pollinated, the seeds were unevenly distributed in the fruit, vertically occupying one side of the fruit. Howpage et al. (1998) described the pollen tube distributor cup as being on the central axis of the ovary and that the cup’s function seems to be in the even distribution of seeds around the fruit core. However, in our results, seeds were not distributed throughout the fruit, but were biased to one side. This indicated that the stigma and carpel had a similar one-to-one like connection. However, the connection was not exact, and seeds were distributed not only in a particular carpel, but also slightly the outside the carpel. Therefore, pollination of the stigmas at a distance is necessary for even seed distribution in the fruit.
Although this study did not reveal the histological relationship between the stigma and the carpel, this uneven distribution of seeds can be inferred from the fact that seeds are formed by pollen tubes extending from a stigma. The control fruit had a significantly larger number of seeds than the one-place treatment fruit. However, even in the one-place treatment, 630 seeds were formed in one fruit in 2018, even though only one or two stigmas were pollinated. The large number of seeds was due to the large number of pollen grains used without dilution. However, even though there were many seeds, the seeds were unevenly distributed on one side, and fruit enlargement was accelerated in the areas with a large number of seeds. On the other hand, in the opposite-end treatment fruit, the fruit enlarged and was well-formed, but the fruit size was smaller than the control fruit.
In kiwifruit, increasing the number of pollinated stigmas increased the seed number. In papaya, a decrease in seed number was associated with a decrease in fruit weight, and an increase in seed number was reported when the number of pollinated pistils increased (Tamaki et al., 2011). It is estimated that approximately 820 seeds or more are required to produce M-size (approx. 100 g) kiwifruit (Fujishima et al., 2006). Kiwifruit weighing 80 g or more are required for market distribution and export (Gonzalez et al., 1998). Together, the results indicate that in order to produce well-formed kiwifruit, the seeds must be evenly distributed. However, the pollination of all stigmas is not always necessary for this. In this study, the number of seeds formed in the one-place treatment ranged from 174 to 631. As the number of seeds was only around 1,000 when all stigmas were pollinated, we predict that the number of seeds would be sufficiently high even if only a small number of the stigmas were pollinated. This finding is expected to contribute to the improvement of artificial pollination techniques in the future.
Regarding the relationship between the number of seeds and sugar content in kiwifruit, Waki et al. (2000) and Fujishima et al. (2006) found that the sugar content increased with the number of seeds, whereas Howpage et al. (2001) and Matsumoto et al. (2007) found no correlation between the two parameters. In grapes, Fujishima et al. (2012) found that the sugar content of seeded fruit was higher than or equal to that of seedless fruit, whereas Xue et al. (2022) reported no difference in the TSS between seeded fruit and seedless fruit at harvest. As demonstrated in those studies, there was no consistent relationship between seed number and sugar content. In this experiment, seed distribution was not related to starch content or TSS. This indicates that seed number affects fruit size more than fruit composition. From previous reports, it is clear that pollination by a large number of pollens is necessary to produce large fruits.
Artificial pollination is an important and common technique in fruit cultivation. However, reducing pollination labor and the amount of pollen used is a longstanding issue. The current study results indicate that the pollination position is important for the production of well-shaped fruits. Further development of agricultural engineering technology should make it possible to pollinate only specific stigmas. This insight into pollination position and pollen quantity should lead to the development of new pollination techniques.
The authors thank Mr. Yukio Koya and Kyohei Takeuchi at the Experimental Farm of the Laboratory of Plant and Animal Science, Meijo University for their assistance and helpful discussions.
