Pollination in an endemic and threatened monoecious herb Begonia satrapis C . B . Clarke ( Begoniaceae ) in the eastern Himalaya , India

Begonia satrapis was studied for its pollination aspects at Sumbuk, Sikkim, India. The floral details and the foraging behaviour of insects visiting the flowers were examined to define the pollination syndrome and its functionality for the success of sexual reproduction in this species. The flowers do not produce nectar and offer only pollen as floral reward to foraging insects. Therefore, male flowers were foraged more for its pollen than the female flowers. There was a significant difference in the visit to male and female flowers by both Apis florea and Bombus breviceps, respectively. The bees spent more time on male flowers than on female flowers. The bees appear to rely on visual stimuli to visit male and female flowers. The plant produces abundant fruit and seed set in both hand and open-pollinations indicating that it is facultatively xenogamous. The female flowers lacking any reward resemble male flowers and in effect are pollinated by deceit.


INTRODUCTION
Male and female flowers provide different levels of rewards to the pollinators. In fact, some female flowers do not produce a pollinator reward and are actually pollinated by deceit (Willson & Ågren 1989;Ågren & Schemske 1991). Deceit pollination can be considered as an extreme case of unreliable signalling in plants since flowers do not offer any reward while they benefit from pollinator visitation (Renner 2006). Intersexual mimicry drives the pollinators to visit the female flowers that do not provide any reward (Little 1983;Ågren & Schemske 1991). Mimicry hypothesis suggests that plants that exhibit intersexual mimicry experience selective advantage when pollinators pay more visits to rewardless female flowers which resemble reward-providing male flowers (Ågren & Schemske 1991). Female flowers in many monoecious species of the genus Begonia may attract pollinators by mimicking conspecific male flowers in which bees mistakenly visit unrewarding female flowers (Vogel 1998;Wiens 1978;Ågren & Schemske 1991;Schemske & Ågren 1995;Corff et al. 1998). This hypothesis is based on the remarkable similarity between unrewarding female flowers and the pollenrich male flowers (Ågren & Schemske 1991). Although, mimicry hypothesis and the behaviour of pollinators have been studied in some species of Begonia (Ågren & Schemske 1991;Castillo et al. 2002;Wyatt & Sazima 2011), no detailed study of such kind has been done in Begonia satrapis C.B. Clarke (Begoniaceae), an IUCN Red Listed Endangered herb of eastern Himalaya (Adhikari et al. 2018). Little (1983) suggested that in floral mimicry hypothesis the mimic and the model floral displays are similar, and the pollinators mostly visit the model more often than the mimic. Accordingly, we tested this hypothesis by studying the floral display of male and female phase inflorescences and the pollinator behaviour in B. satrapis. The study addressed the following questions: (i) are there any differences between the size and morphology of male and female flowers? (ii) do pollinators discriminate between male and female flowers? (iii) what is the success rate of intersexual mimicry to deceive the pollinators to effect pollination?

Study site
This study was conducted in a private forest in Sumbuk which falls in the Rangit Valley, South District, Sikkim, eastern Himalaya (27 0 06'18.90"N & 88 0 22'07.32"E, altitude 555m). The area experiences a maximum and minimum temperature of 26.9 0 C and 17.3 0 C, respectively, with an annual precipitation of 2,766mm. The study site comprises a sub-tropical type of forest where B. satrapis flourished in abundance along the margins of this forest which is close to human habitation. The forest surroundings comprised species of Shorea robusta C.F. Gaertn and Schima wallichii Choisy.

Statistical data analysis
Mann-Whitney U Test was performed to evaluate the difference in seed set rate in hand and open pollination of female flowers. Non-parametric t-test was used to evaluate the variation in morphological characters of male and female flowers. Data collected from different patches were pooled and subjected to a t-test to know whether the resulting variation levels are statistically significant or not. A t-test was performed between the open flowers and closed buds of both male and female flowers, respectively, to check if there is any chronological difference in the opening of male and female flowers.

Inflorescence sex ratio and floral morphology
A sample of 50 flowers, each for male and female sex, was used to record floral morphometrics. A sample of 50 plants was used to record the average number of male and female flowers produced in individual inflorescences. Anthesis schedule and flower lifespan were observed in the field itself.

Foraging activity and pollination
Pollinators were observed in three 3×1 m randomly chosen patches of B. satrapis which were reselected on 31 August, 1, 2 and 14 September 2017 (as per Ågren & Schemske 1991). The observation period for the day continued until the pollinators ceased to visit the flowers. To test if pollination occurs by wind, several of the buds (N=25) were bagged with mosquito nets which were made into small bags to exclude visitors in order to record the fruit and seed set rate. Similarly, another set of buds (N=25) were bagged and tagged with ribbons to test apomixis.
Apis florea and Bombus breviceps were the most frequent pollinators throughout this study. A few rare visits by two other unidentified bee species were noted. The observations were made only on A. florea and B. breviceps as other foragers visited the flowers rarely.

Hand pollination
Female buds (N=50) were bagged and tagged with a coloured ribbon a day before their anthesis and were hand pollinated on the following day. Hand pollination was performed by brushing the fresh dehisced anther against the stigma of a different plant. Similarly, female buds (N=50) were tagged with a different coloured ribbon and were left for open pollination. The matured fruits were collected to record the number of seeds produced against the number of ovules produced per flower. The pollen limitation was estimated as the ratio of hand cross pollination to open pollination (Larson & Barrett 2000). The scale ranged from 0-100 where 0 indicated no pollen limitation to 100 indicating pollen limitation. All the tagged flowers could not be retrieved at the time of their collection due to anthropogenic activities at the study site.

Floral morphological details
The flowers of B. satrapis bloom during July-October. The flowers are open from 05.00-06.00 h. A female flower lasts for 7-9 days while a male flower lasts for almost 15 days. The inflorescence is a cymose with male and female flowers with pink tepals (Image 1). Male flowers comprise four tepals (2+2) while female flowers comprise five tepals (2+3). The outer tepals of male flowers were significantly longer (1.5 ± 0.24 vs. 1.2 ± 0.17 cm; t=1.55e-09, p<0.05) and wider  (Table 2). Apis florea used their legs to remove the pollen while B. breviceps performed vibration to collect pollen. Both bee species discriminated female flowers after making first visit to them. The foraging activity of A. florea was slower than B. breviceps while there was no significant difference in time spent on male flowers (t=0.31, p>0.05, NS) and female flowers(t=0.13, p>0.05, NS) by both bee species.

Hand pollination
Although fruit set was observed in both hand pollinated and open pollinated flowers, hand pollination between male and female flowers of different plants resulted in an increase in seed set (95.4 ± 8.1, N=17) than open pollination (81.1 ± 17.2, N=17) and showed a significant difference between the two (U=0.001, p<0.05). Fruit and seed set was absent in both wind pollination and apomixis.
The estimated pollen limitation (1.17) indicated that the pollinators deposit adequate pollen in its natural environment.

DISCUSSION
B. satrapis is a monoecious species with both male and female flowers borne in the same inflorescence whose morphological characters varied significantly between male and female flowers in a population. The larger tepals of male flowers appear to be an adaptation to provide visual stimuli to pollinator bees to locate the flowers that provide the reward (Ågren & Schemske 1991). The anthesis of male flowers prior to female flowers enable the pollinators to habituate themselves to the forage source and visit rewardless female flowers by deceit when available on the same or different inflorescences of the same or different conspecific plants. Similar observations were reported in Jacaratia dolichaula (Bawa 1980) where pollinators first encountered the rewarding male flowers and therefore, reduced the chances of discrimination of early flowering rewardless female flowers (Corff et al. 1998). The stigmas of female flowers are yellow and strongly resemble male flowers. The female flowers attract pollinators by mimicking male flowers (Wiens 1978;Ågren & Schemske 1991). The rimose anthers are grouped in large number which facilitate pollen collection by vibration (Wyatt & Sazima 2011). The two important foragers A. florea and B. breviceps showed more preference to the male flowers than the female flowers. This possibly could be because of the pollinators ability to recognize unrewarding flowers (Wyatt & Sazima 2011). It was observed that A. florea seemed diffident to collect pollen from a male flower immediately after its visit to a male flower. The lower rate of visitation and the hesitation shown in collecting pollen could be because of the lack of fragrance which has been experimentally proven to be an important aspect to encourage landing on a female flower (Lunau 1991;Schemske & Ågren 1995). Despite the negligible visits to a female flower a high seed set was observed in flowers left for open pollination which could be because of pollination by vibration (Wyatt & Sazima 2011). When a pollinator performs vibration while it visits a female flower by deception it transfers a large load of small and powdery pollen to the stigma (Wyatt & Sazima 2011). A single visit is adequate to deposit a large amount of pollen load to the stigma (Wyatt & Sazima 2011). Seed set in hand pollination was significantly higher than that of open pollination. The lack of pollen limitation, however, indicates that the pollinators are efficient in depositing pollen for successful seed set. Since wind pollination and apomixis are absent in B. satrapis, it becomes evident that it depends largely on pollinators for its successful sexual reproduction.

CONCLUSION
It is apparent from our study that the pollinators prefer to visit the polleniferous male flowers more than the rewardless female flowers. The imperfect discrimination by the pollinators by the number of   www.threatenedtaxa.org The Journal of Threatened Taxa (JoTT) is dedicated to building evidence for conservation globally by publishing peer-reviewed articles online every month at a reasonably rapid rate at www.threatenedtaxa.org. All articles published in JoTT are registered under Creative Commons Attribution 4.0 International License unless otherwise mentioned. JoTT allows allows unrestricted use, reproduction, and distribution of articles in any medium by providing adequate credit to the author(s) and the source of publication.