Background and Aims Winter-flowering plants outside the tropics may experience a shortage of pollinator service, given that insect activity is largely limited by low temperature. Birds can be alternative pollinators for these plants, but experimental evidence for the pollination role of birds in winter-flowering plants is scarce.
Although the most common use buying sildenafil for a penis pump and let you enjoy great sex without being troubled by ED. Certain bacteria are harmful for our body and attack our body when got exposed to the infected environment. tadalafil buy india Erectile dysfunction is an inability to sustain an erection and maintaining an erection is difficult. cheap levitra online Foods vardenafil levitra online rich in zinc include wheat germ, oysters, milk full fat, chick peas, rye, lamb, crab cooked, roast beef, chicken breast, cheese, fish, cooked brown rice, low fat yogurt, canned baked beans, cashews, pine nuts, pumpkin seeds, shrimp, almonds, pecans, brazil nuts, sesame seeds, almonds, spinach, kidney beans, flax seeds, lima beans, garlic, watermelon seeds, salmon, turkey, lobster, dark chocolate and mushrooms.
Methods Pollinator visitation to the loquat, Eriobotrya japonica (Rosaceae), was observed across the flowering season from November to January for two years in central China. Self- and cross-hand pollination was conducted in the field to investigate self-compatibility and pollen limitation. In addition, inflorescences were covered by bird cages and nylon mesh nets to exclude birds and all animal pollinators, respectively, to investigate the pollination role of birds in seed production.
Results Self-fertilization in the loquat yielded few seeds. In early winter insect visit frequency was relatively higher, while in late winter insect pollinators were absent and two passerine birds (Pycnonotus sinensis andZosterops japonicus) became the major floral visitors. However, seed-set of open-pollinated flowers did not differ between early and late winter. Exclusion of bird visitation greatly reduced seed-set, indicating that passerine birds were important pollinators for the loquat in late winter. The whitish perigynous flowers reward passerines with relatively large volumes of dilute nectar. Our observation on the loquat and other Rosaceae species suggested that perigyny might be related to bird pollination but the association needs further study.
Conclusions These findings suggest that floral traits and phenology would be favoured to attract bird pollinators in cold weather, in which insect activity is limited.
Flowers exhibiting the ornithophilous syndrome show certain characters (of morphology, colour, nectar and odour) that presumably attract birds as effective pollinators. The typical traits of ornithophily include showy tubular corollas, often bright reddish, odourless, with diurnal anthesis and copious nectar (Stiles, 1981; Proctor et al., 1996; Pellmyr, 2002). Normally, the ornithiphilous syndrome is associated with three groups of birds (hummingbirds, honey-eaters and sunbirds) (Ortega-Olivencia et al., 2005; Cronk and Ojeda, 2008) distributed in the New World, Oceania and the Old World, respectively. Most studies on bird pollination have been carried out in the New World (Feinsinger, 1976; Stiles, 1981; Nattero and Cocucci, 2007; Martén-Rodríguez et al., 2010), where hummingbirds were the main pollinators.
Bird pollination is considered to be almost entirely absent in Europe and in Asia north of the Himalayas (Proctor et al., 1996; Cronk and Ojeda, 2008). Although anecdotal observations showed that passeriforms feed on nectar in numerous plants in these regions (e.g. Stiles, 1981; Proctor et al., 1996;Harrup, 1998; Schwilch et al., 2001; Merino and Nogueras, 2003; Corlett, 2004; Valido et al., 2004), experimental evidence of bird pollination is rare. In Japan, white-eyes (Zosterops japonica) were reported to be the most effective pollinator for Camellia japonica during winter months when insects are few (Kunitake et al., 2004). The white-eyes and their relatives (Zosteropidae) are small, omnivorous birds having brush-tipped tongues that may facilitate nectar sucking, as in honeyeaters (Corlett, 2004). The white-eyes have also been observed pollinating Taxillus yadorigi(Loranthaceae) in the warm temperate forest (Yumoto, 1987) andRhododendron barbatum in western Japan (Huang, 2011). These bird-pollinated plants usually have large, open-shaped, red flowers with large volumes of nectar. However, three warblers were confirmed as pollinators for Anagyris foetida L. (Fabaceae), a native tree with yellowish flowers in Spain (Ortega-Olivencia et al., 2005). It has been suggested that passerine birds were potential pollinators for certain flowers with abundant nectar, despite the fact that these flowers did not exhibit the showy colours typical of the ornithophilous syndrome (Ford, 1985).
Plants flowering in winter are usually faced by insufficient pollinators, given that low temperature in winter restricts the activity of insect pollinators. For example, honey-bees become immobile when the air temperature is below 7–10 °C. Many generalized pollinators, such as bumble-bees, hibernate during winter (Goulson, 2003). As an alternative pollination mode, bird pollination was believed to be more dependable where cold and/or rainy weather conditions are frequent, as on high mountains or in Mediterranean-climate winters (Stiles, 1971, 1978;Cruden, 1972). In Israel, Loranthus acaciae (Loranthaceae) flowering in summer was visited by a broad spectrum of pollinators, while that flowering in winter was only visited by the Palestine sunbird (Nectarinia osea) (Vaknin et al., 1996).
The loquat, Eriobotrya japonica (Rosaceae), is an evergreen fruit tree that flowers in winter. It is indigenous to south-eastern China (Zhang et al., 1990). The loquat flower is yellowy white, bowl shaped and with nectar, presenting a generalist pollination syndrome. As an early fruit that can be harvested in spring, it has been cultivated in the Mediterranean region and Central America. In the Mediterranean region, honey-bees (Apis mellifera) were the most important pollinators, while other bees in the generaAnthophora, Xylocopa and Halictus were also observed (Freihat et al., 2008). In the United States, loquats have been recorded as pollinated by diverse insects, including bees, syrphids, houseflies, Myrmeleontidae, Bombinae and Pieris rapae (Crane and Caldeira, 2006). In Spain, four bird species (Phylloscopus collybita, Sylvia melanocephala, Sylvia atricapilla andParus caeruleus) were observed to probe nectar from E. japonica (Merino and Nogueras, 2003), but whether these birds were effective pollinators in this species remains unclear. In China, the loquat native area, the pollinator composition is relatively unknown. Low temperature at loquat flowering time might restrict insect visitation. Here we have demonstrated experimentally that passerine birds serve as alternative pollinators for E. japonica when the temperature is low in winter. The finding of resident passerines functioning as pollinators at low temperature may provide an explanation for winter or early-spring flowering of numerous trees with perigynous flowers in the rose family, Rosaceae.
MATERIAL AND METHODS
The study species
Eriobotrya japonica originated in the provinces of Sichuan, Chongqing and Hubei, south-east China (Zhang et al., 1990), and was widely cultivated throughout Southeast Asia and in other countries. The inflorescences constitute stiff panicles 10–19 cm long and contain 60–100 flowers covered by a thick yellow–brown pubescence (Fig. 1). The flowers are fragrant, 1·2–1·7 cm in diameter, with no fixed open direction, yellowy-greenish when young but turning to yellowy-white when flowering. The androecium consists of about 20 free stamens. The stamens curl inwards when young, unbending progressively from outside to inside. At the base of the calyx the perigynous flower has nectaries whose nectar is deposited onto the shallow hypanthium. Our field survey was conducted from November 2009 to March 2010 and November 2010 to January 2011 in the campus of Wuhan University, Wuhan, Hubei province (32°32′N, 114°21′E). The population consisted of about 40 transplanted flowering individuals at 30–50 m a.s.l. on the south side of Luojia Hill within the native range of the species. It is characterized by a subtropical monsoon climate of chilly humid winter with an average temperature of 6 °C, and 1250 mm annual precipitation (Zhou et al., 1999). Most individuals bloomed in December and January. During winter, the inflorescence was sometimes covered by snow (Fig. 1A).
Inflorescences, flower visitors and infructescences of Eriobotrya japonica. (A) A flowering inflorescence covered by snow; (B) a section of the flower showing nectar droplets; (C) a honey-bee visiting caged flowers from which birds were excluded; (D)Pycnonotus sinensis visiting flowers; (E) Zosterops japonicusvisiting flowers; and (F) fruits in April.
Pollinator observations
To quantify pollinator visiting frequency to the loquat, we conducted systematic bird and insect censuses under diverse weather conditions from November 2009 to March 2010 and November 2010 to January 2011, totalling 95 and 65 h, respectively. In the first year, we separated the observation time into five periods, each with 100·0 ± 17·7 censuses. In the second year, we separated the observation time into four periods, each with 97·5 ± 19·8 censuses. Each census lasted 10 min for insects or birds. In insect censuses, observers were about 2 m from the inflorescences, permitting us to discriminate visitors probing for nectar or pollen. In bird censuses, observers were at a distance of 2–10 m, observing through a long-focal-length camera. We only recorded birds that probed nectar or contacted anthers. Visitation frequency was obtained as the mean number of visits per flower for each visitor species. Because the birds were in flocks, their visitations were pulsed. We calculated the average number of bird visits per flower by multiplying the number of visits made by one haphazardly chosen bird by the number of birds in the group, and dividing by the total number of flowers on the observed inflorescences. We used SPSS 13·0 to conduct standard statistical tests.
Breeding system
To determine the breeding system of E. japonica in the population, we bagged 30 inflorescences with nylon mesh nets prior to the start of flowering, and randomly assigned plants to cross- and self-artificial pollination. Loquat is considered a self-compatible species, while cross-pollination improves fruit-set (Crane and Caldeira, 2006). Therefore, in the cross-pollination treatments anthers were removed from the flowers before they dehisced and the flowers received pollen obtained from randomly selected plants approximately 20 m away. Fruits were collected the following spring, 3 months after pollination. Data for seed-set under the self-pollination treatment were not normally distributed, so they were analysed using non-parametric statistics (Mann–Whitney test). We randomly collected 20 opened flowers with undehisced anthers to count pollen grains and ovules per flower (Tang and Huang, 2007).
Examination of bird pollination
To examine the pollination role of birds, inflorescences were randomly selected for one of three treatments: (1) exclusion of all pollinators using small-mesh (1 × 1 mm) nylon nets; (2) exclusion of large pollinators such as birds using iron cages with intertwined cotton threads with large mesh (15 × 15 mm) (Fig. 1); and (3) open pollination (open to all floral visitors). Fruits under these treatments were collected and seeds per fruit (ranging from 1 to 10) were counted. We tested for differences in seed-set among the experimental exclusions by one-way ANOVA and Tukey’s post-hoc tests (variances being homogeneous, Levene’s statistic F = 2·876, P = 0·062).
Nectar volume and concentration
During the study period, we sampled 20 flowers on four occasions to estimate the production of nectar from December to January. The flowers were bagged with small-mesh nylon bags for 1 day and the nectar was extracted by micropipettes. Sugar concentrations were then measured by pocket refractometers (Bellingham + Stanley Ltd, Tunbridge Wells, UK; 0–50 %).
RESULTS
Six types of insect species, butterflies (Vanessa cardui), honey-bees (Apis cerana), wasps (Vespa velutina), flies (Calliphora vicina), hoverflies (Episyrphus balteatus, Eristalis cerealis, Eristalis tenax) and hawkmoths (Macroglossum stellatarum), were observed visiting the loquat flowers (Fig. 1). In the first year, honey-bees were the most frequent floral visitors, accounting for 91·7 % of the total (2721) insect visits. Others included 181 visits by hoverflies (6·1 %), 59 visits by flies (2·0 %) and seven visits by hawkmoths (0·2 %). In the second year, a wasp (Vespa velutina) became the most frequent visitor, accounting for 45·0 % of the total (1275) insect visits, followed by 24·6 % visits by hoverflies and 22·7 % by honey-bees. Honey-bees and wasps were observed to forage for nectar and rarely for pollen in most visits although flowers offered plenty of pollen. Hoverflies usually settled on the anthers, feeding on pollen, but rarely contacted the stigmas, given that they could not access nectar through the fence constructed by the filaments.
The major bird pollinator was the Chinese bulbul (Pycnonotus sinensis, Fig. 1D; also called the light-vented bulbul), which accounted for 95·1 % of bird visits (2795) and 100 % of bird visits (125) in two study periods, respectively. White-eyes (Zosterops japonicus, Fig. 1E) were also observed to pollinate flowers, accounting for 4·9 % of visits (144) in the first year. When the two bird species inserted the beak into flowers, pollen grains were seen attaching to the beak and face. Insect activity was restricted by the daytime temperature. Insect visit frequency was positively related to the average temperature (Spearman rank correlation, coefficient = 0·57, P= 0·02). In November, honey-bees were active from 0900 to 1600 h on sunny days. In December, the activity phase of honey-bees was shortened, lasting from 1000 to 1500 h. Insect visit frequency decreased in January, whereas bird visitation increased (Fig. 2). No insects were observed to visit loquat flowers in late January. In the second year, bird visit frequency was low. However, bird visits were observed during late January in both years when air temperature was lowest, suggesting that birds contributed greatly to pollination in cold weather. Consistent with this result, we observed that seed-set of inflorescences that flowered in January was slightly higher (2·9 ± 1·5 %, mean ± s.e.) although it was only marginally significantly (P = 0·08) different from that of flowers in November (2·2 ± 0·9 %).
Visit frequency (mean ± s.e.) of insects and birds during observation periods in winter 2009 and 2010.
Hand-pollination experiments indicated that E. japonica was self-incompatible; mean seed-set under self-pollination was very low (0·022 ± 0·012). Seed-set of open-pollinated flowers (0·026 ± 0·002) was significantly higher than that resulting from selfing (Z = –6·25, P < 0·01) but was significantly lower than that resulting from cross pollination (0·193 ± 0·026; Z = –2·33, P < 0·01). The results of hand-pollination treatments indicated that seed production in the loquat was pollen limited and could be promoted by cross pollination.
Insect visitation frequency did not differ (two-tailed t-test; t = –1·37, P = 0·25) between the bird cage and open pollination treatments, indicating that large-mesh cages did not restrict insect visitation. Seed-set did not differ between treatments that excluded birds with bird cages and those that excluded all pollinators with mesh nets, but both were lower than that of open-pollinated inflorescences (Fig. 3). These results indicated that insects did not play an important role and birds participated in pollination. By the exclusion of bird pollination, seed-set was significantly reduced to 50 % (Fig. 3).
Seed set (mean ± s.e.) inEriobotrya japonica under three pollination treatments: cages excluding birds; mesh excluding birds and insects; and unprotected controls. Different letters above the boxes indicate significant differences between treatments (one-way ANOVA, Tukey HSD, cage vs. mesh, P = 0·36; cage vs. control, P < 0·01; mesh vs. control, P< 0·01). Numbers within the boxes indicate treated inflorescences.
Anthesis generally lasted 4–5 d in early winter (November) and 6–7 d in late winter (January). One flower produced 95361 ± 5244 (mean ± s.e.) pollen grains and ten ovules. The accumulated nectar (over 40 µL) often filled the hypanthium and submerged the unopened stamens if flowers were unvisited. The mean volume of nectar per flower after bagging for 24 h was 12·30 ± 3·72 µL, and the mean sugar concentration was 12·01 ± 2·62 %. Volumes and sugar concentrations varied over the flowering periods (one-way ANOVA, F = 8·53, P < 0·01; F = 18·32, P < 0·01, respectively), but for sugar content did not (F = 0·25, P = 0·86). Flowers sampled in January had larger volumes (14·2 ± 3·3 µL; two-tailed t-test, t = –4·68, P < 0·01) of more dilute nectar (10·5 ± 1·9 %; t = 5·30, P < 0·01) than those sampled in December (10·7 ± 3·3 µL, 13·3 ± 2·5 %).
DISCUSSION
Our study demonstrated that passerine birds served as effective pollinators in Eriobotrya japonica, a winter-flowering fruit tree in Central China. In the studied population, bees were the major visitors in early winter when the temperature was relatively high, and birds were major visitors in late winter when the temperature was relatively low. Seed-set was similar in the early- and the late-flowering seasons, in which E. japonica experienced different types of pollinators, suggesting that loquat flowers could be pollinated by both bees and birds, a mixed bird–insect generalist pollination system (see Ollerton et al., 2009b). Such mixed pollination system involving bees and birds was also observed in Aechmea nudicaulis (Bromeliaceae) in the Atlantic rain forest of southern Brazil (Schmid et al., 2011). Our exclusion of bird pollination significantly decreased seed production, confirming the pollination role of birds in this fruit tree.
Other observations outside Asia showed that honey-bees were the major pollinators for loquat in the Mediterranean (Freihat et al., 2008) and Florida, United States (Crane and Caldeira, 2006), while birds were observed visiting flowers in Spain (Merino and Nogueras, 2003). Flower visitors to E. japonica in our studied population were diverse and variable. Bees dominated in the early winter, primarily searching for nectar instead of collecting pollen. We did not observe obvious pollen loads on the hind legs of honey-bees. Perhaps winter is not a time for honey-bees to collect pollen for raising broods. In comparison, passeriforms tend to forage and travel in flocks and can be effective in cross-pollination even for large trees (Stiles, 1981). Our hand-pollination treatments indicated that the species was self-incompatible rather than self-compatible (Crane and Caldeira, 2006). This difference needs further study. One possibility is that different varieties evolve differential levels of self-compatibility. Among bird visitors, the Chinese bulbul appeared to be the primary bird pollinator for E. japonica given that it is resident whereas the white-eye is partly migratory at this latitude. The Chinese bulbul flocks that we observed normally comprised over 40 individuals, and were generally larger than white-eye flocks, which contained fewer than 20 individuals. The Chinese bulbul was observed to consume petals and stamens in E. japonica, which might damage the flower’s reproductive organs during visitation (Merino and Nogueras, 2003). However, the bulbul seems unlikely to affect seed-set given that inferior ovaries are enclosed by the receptacle in the perigynous flower.
The major flowering period was from December to January of the following year. The shift between insect and bird pollination depended on the time of the first snowfall. In the study site, the average temperature over 24 h in December and January was 5·1 °C, taking the two years together. In 70·9 and 77·4 % of days, the average temperature was below 7 °C (threshold of honey-bee activity temperature) during the October 2009 and November 2010 study periods, respectively. Low temperature restricted the activity of insects. It appears that this nectar-feeding habit during winter enables Chinese bulbuls to be effective pollinators of E. japonica despite their wide-ranging diet over the remainder of the year.
Although flowers of E. japonica are pale and fragrant, not typical traits of ornithophily (see Ford, 1985; Ollerton et al., 2009a), their attractiveness for birds is due to the copious production of nectar and long-lasting flowering period. Nectar sugar concentration was similar to that of bird-pollinated flowers in southern Africa (10–15 %), but more dilute than that consumed by hummingbirds (20–25 %) (Nicolson, 2002) and by bees (35–40 %) (Harder, 1986). The relatively large volume, dilute nectar and inferior ovary are characteristics of ornithophilous species (Nicolson, 2002; Cronk and Ojeda, 2008). The open-shaped flower in E. japonica cannot protect nectar from evaporation. However, the inner curled stamens and continuously humid weather reduced the rate of nectar evaporation, stabilizing the quality of the nectar reward. Both bees and birds could be effective pollinators sucking nectar from the loquat. Given the well-documented association of nectar sugars with pollinator groups (Dupontet al., 2004; Cronk and Ojeda, 2008), it would be interesting to know where the nectar is sucrose- or hexose-dominated. The mixed bird–insect pollination system in the winter-flowering E. japonica may be an adaptation to severe temperature dynamics in winter. Temperature in winter in non-tropic Asia is lower than the threshold for insect activity for most of the time. Therefore, visits by insect pollinators become infrequent and unpredictable in changeable winter temperatures. By contrast, birds are warm-blooded, becoming more reliable pollinators at low temperatures where cold and/or rainy weather conditions are frequent (Stiles, 1971).
Omnivorous birds as effective pollinators have rarely been reported in Europe and in Asia outside the tropics (Kunitake et al., 2004; Ortega-Olivencia et al., 2005; Zhang et al., 2011). During its long flowering period (approximately 4 months for the whole population; 2 months per plant; 5–8 d per flower), E. japonica is available for visitation by various pollinator species. The flowering season of E. japonica spans the coldest period in central China. This could reduce competition for pollinators and heterospecific pollen interference, given that other flowering plants are very scarce in this season. Also, flowering in winter could reduce pre-dispersal seed predation (Brody, 1997). The phenological concordance between the resource shortage for pollinators and the flowering of E. japonica may have led to the seasonal nectar feeding of Pycnonotus sinensis, and as a result P. sinensis provides an alternative effective pollination service in winter.
The perigynous flower, a relatively infrequent floral form in angiosperms, is common in several genera of Rosaceae, in which the receptacle grows up around the ovary into a kind of cup, for example Chaenomeles,Eriobotrya, Malus and Prunus. Grant (1950) noted a close association between perigynous flowers and beetle pollination and suggested that perigyny may represent an adaptation in beetle flowers for ovule protection from beetle damage. Passerines were observed sucking nectar from numerous species with a floral cup producing nectar in winter or early spring, such as Chaenomeles speciosa, Prunus mume, Prunus serrulata and Prunus persica (S.-Q. Huang, unpubl. res.). Our observation on winter-flowering Eriobotrya japonica and other species of Rosaceae suggested that perigyny might also be related to bird pollination, given that it can protect the ovules from the bill of the bird (see Grant, 1950) as well. Clearly, the association of perigyny with bird pollination needs further study in other species, such as in Rosaceae.