Flower Pollinator Interactions within Two Tropical Tree Species of Mizoram , North East India

Understanding the interactions between plants and pollinators is vital in exploring the structural and functional dynamics of tropical forest ecosystem. Such interactions are also helpful in resolving issues of regeneration, maintenance of biodiversity and conservation of tropical forest resources. The aim of this study was to understand the foraging behaviour of floral visitors of two valuable forest species in North East India, viz. Schima wallichii Choisy and Lagerstroemia speciosa (L.) Pers. The types and behaviour of pollinators were assessed by counting of the total number of visits by insects during day time (05:00-18:00) per field visit/flower in the peak flowering season, foraging rate (number of flowers visited/minutes) and average time spent per flower. The frequency of pollinator visits was also determined in terms of visits/flower/hour. Ten species belonging to seven families were recorded visiting S. wallichii flowers. The flowers of L. speciosa were mainly visited by three species in the family Apidae and one species in the family Calliphoridae. Various moth and bee species were involved in pollination of S. wallichii flowers. The Xylcopa spp. of bee showed a high level of functional synchronisation with the floral morphology of


Introduction
Plant-pollinators interactions are one of the key ecological steps in the reproductive success of plants, particularly in outcrossing species.There has been heightened interest in tropical pollination biology in recent decades due to the phenomenon of global environmental change (Bond, 1994;Murren, 2002).Tropical forest ecosystems are facing accelerated rates of habitat conversion to other land-use systems, leading to loss of biodiversity (Vitousek, 1994).Furthermore, there is a concern regarding the failure of mutualistic interactions (plant-pollinator interaction) due to habitat destruction and fragmentation, which may lead to the cascade extinction of species and loss of valuable biodiversity (Bawa, 1990;Kearns and Inouye, 1997;Renner, 1998).Tropical plant species with disperse and patchy distribution, self-incompatibility, dioecy and high level pollinator specificity are more vulnerable to extinction in case of failed pollination (Bond, 1994;Corlett, 2004).Tropical forest tree species offer a unique opportunity to explore evolutionary tendencies in relation to morphological diversification.Understanding the interactions between plants and pollinators is vital in exploring the structural and functional dynamics of tropical forest ecosystem.Such interactions are also helpful in resolving issues of regeneration, maintenance of biodiversity and conservation of tropical forest resources (Bawa and Headley, 1990;Sharma et al., 2010).
Understanding the diversity and behaviour of floral visitors to different species from a range of ecological characteristics is helpful in evaluating the features of different pollination systems.Key features are: species relationship (Fenster et al., 2004), specific evolutionary tendency between plants and animals (Thompson, 1988;Mitchell et al., 2009) and the structural and functional dynamics of a particular community (Memmott, 1999).Flower anthesis provides a new resource for local floral visitors.Also, local fauna may visit opened flowers for a variety of reasons, such as to feed on nectar or pollen (Bawa, 1990), for shelter (Sakai, 2000) and/or to predate upon other floral visitors.Pollinators are vital in successfully transferring pollen to conspecific individuals at the population level and govern the level of gene flow (House, 1998).Scientific studies are mainly focused on direct observations to record, identify and quantify floral visitors (Momose et al., 1998).For example, even flowers that appear to require specialised pollinators are often visited by a diverse array of faunal visitors (Johnson and Steiner, 2000).Changes in climate are raising concerns about the size of pollinator population and their ecological services (Hegland et al., 2009).Therefore, it is necessary to determine the current profile of floral visitors to flowering plants of agricultural and forestry importance (Bronstein 1995;Waser et al., 1996).
Studies of pollinators' diversity of forest genetic resources from the North-East India are lacking.The region also covers biodiversity rich habitat of flora and fauna.The rate and quantity of pollinators' visits to plant species is frequently linked with reproductive success through pollen dispersal (Burd, 1994).Different floral visitors disperse pollen across varying distances, thus affecting pollen flow.Exploring the assemblage of pollinators would be helpful in making conservation and management decisions, as reproduction of animal pollinated plants and their demographic growth is significantly influenced by the reduction in pollinator availability (Kearns et al., 1998).Perturbation in pollinator communities may not only negatively impact seed and fruit set, but also may reduce the adaptive fitness of progeny through changes in the amount and composition of pollen loads (Burd, 1994).Many studies recently revealed that selfing rates are driven by spatial and temporal variation in the composition and abundance of native pollinating animal visitors (Harrison, 2000;Fenster et al., 2001;Christopher et al., 2003;Mitchell et al., 2009).Therefore, it is important to understand the range of pollinator species and their behaviour with regard to valuable forestry trees.
The tropical forest ecosystem of North East India is extremely rich in floral and faunal diversity, but studies on functional aspects of biodiversity are lacking from this region.In general, there is paucity of basic information on the species of pollinators involved in pollination of valuable forest genetic resources.Therefore, the present study was undertaken to map and understand the foraging behaviour of floral visitors of two valuable forest species, Schima wallichii and Lagerstroemia speciosa.

Study site
The study site was in and around Mizoram University Campus, Tanhril, Aizawl, which is situated about 15 km West of the state capital Aizawl, just below Tanhril village.The study area lies in between 23 0 45'25'' and 23 0 43' 37" N latitudes and 92 0 38' 39" and 92 0 40' 23" E longitudes, with the elevation ranging from 300 m to 880 m asl.The area is characterised by a series of undulating slopes with the Western spur fallings under steep slope of the bank of Setlak River.The climate is humid and tropical, characterized by a short winter and long summer with heavy rainfall (2,100 mm).The temperature did not fluctuate much throughout the year, and ranged from 12 °C to 36 °C.

257
Schima wallichii Schima is a monotypic, polymorphic genus belonging to the family Theaceae that has a single species, Schima wallichii Choisy.It is an evergreen tree native to warm temperate and tropical regions of Southern and South-Eastern Asia.It is locally used for timber, poles, fuel wood, charcoal making and occasionally as fodder (Tamrakar, 1992).Flowering starts at the age of 5-8 years, depending on growth and site.Seasonal flowering occurs in April-early June with most seeds maturing in January to March next year.The development from flower to fruit thus takes about 9 months.The tree attains a height up to 35 m with a dense crown and can attain a stem diameter up to 125 cm.Leaves are elliptic-oblong or lanceolate, 6-15 × 4-6 cm in size; flowers are white, 2.5-4 cm across, scented, axillary solitary or paired.It has a broad distribution occurring from North-Eastern India to Southern China, Thailand, the Malay Peninsula, Sumatra, Java, Borneo and the Philippines, and is adaptive to a wide range of environments (Corlett, 1999;Wu et al., 1999).S. wallichii is a fire resistant, moderately light demanding and often occurs in diverse primary lowland to mountain forests, scrublands and grasslands, but is particularly common in disturbed and secondary forests (Troup, 1921;Li-Zhen and Li, 1997).Bark contains alkaloids that are used as fish poison and also used for dying (Li-Zhen and Li, 1997).Corollas of the flowers are used to treat uterine disorders and hysteria (Tamrakar, 1992;Siwakoti and Siwakoti, 2000).

Lagerstroemia speciosa
Lagerstroemia speciosa is a large deciduous tree that grows up to 18 m in height with a diameter of up to 50 cm.Leaves are elliptic-lanceolate.Flowers are mauve coloured in large terminal peduncles up to 25 cm long.Each flower is 5-8 cm in diameter with a calyx of 12-15 mm long, tomentose and ribbed with 6 teeth.There are 6-7 petals per flower, each 3-4 cm long (Sinha et al., 2012).Flowers are produced once a year in the height of summer.It is commonly known as Queen's Crape-Myrtle due to its conspicuous beautiful flowers.The species grows naturally in North-Eastern and Southern India, throughout Myanmar, the Philippines and the Chittagong hills of Bangladesh (Troup, 1921).It is a multi-purpose tree species with good quality timber and medicinal values (Suzuki et al., 1999).Seeds are narcotic; the bark and leaves are purgative; the roots are astringent, stimulant and febrifuge (fever removing).A decoction of dried leaves is used in diabetes (Mishra et al., 1990) and anti-obesity (Suzuki et al., 1999).

Floral, pollination and fruit display
This study was conducted during the flowering seasons (mid April-May for S. wallichii and mid May-June for L. speciosa) of the years 2012 and 2013, which covered about one month period for each species in every year.Floral morphometric traits, pollination and fruit formation were observed during the flowering season on ten randomly selected individuals within each population.On each tree, twenty flower buds of similar stage were marked and subsequently observed for anthesis, anther dehiscence and stigma receptivity at 2-h intervals from dawn (05:00 h) to dusk (18:00 h).Anthesis (the opening of flowers to display its reproductive sexual organs) was observed on the marked flowers on each tree of both study species.Anther dehiscence was assessed at the time of anthesis by using a hand lens (×20) to note the time differences between anthesis and anther dehiscence, and anther dehiscence and stigma receptivity.Receptivity of stigmas was assessed up to 5 days after every 2-h interval during entire day length, by measuring peroxidase activity using the method of Kearns and Inouye (1993).Furthermore, twenty flowers from different trees were harvested to record the flower diameter, sepal and petal length, style, ovary length and ovary circumference, for each species.Similarly, 50 mature fruits from different selected individuals of each species were harvested and the wing length of seeds, seed length (without wing), number of seeds per fruit, number of fruits per inflorescence, length of fruit and circumference of fruit were measured.

Pollinator availability and their frequency
Ten individuals of each study species (S. wallichii and L. speciosa) were chosen arbitrarily within the populations to record the pollinators.Pollinators and their visitation rates were determined on five randomly chosen branches per tree by counting the individuals of insect species visiting to the selected branches per individual tree of each species.Thus, the sample size represented 50 replications (10 × 5 = 50) in each study species.Pollinator availability was monitored during peak flowering time (mid April-May for S. wallichii and mid May-June for L. speciosa).Several inflorescences were observed simultaneously to record the number of encounters of pollinator species interacting with open flowers.Each tree was observed for whole day between 05:00 and 18:00 h in 6-h observation blocks, at every full hour interval between 05:00-06:00, 06:001-07:00, 08:01-09:00 h, etc.The observations were recorded for twenty five days during peak flowering time of both species.The sustainable distribution of each observation block was done in such a manner that all the chosen trees in the population would be monitored in each observation block during the course of the study.The pollinators were assessed in respect of total number of visits by insects during field visit/flower (n = 10 trees × 5 branches × 10 flowers per branch = 500 flowers), foraging rate (number of flowers visited/minutes, n = 500) and average time (sec) spent per flower (n = 500).The frequencies of pollinators were also assessed in terms of visits/flower/hour (n = 500).

Results
The visitation rates differed significantly (p < 0.0001) within the insect species and also varied significantly (p < 0.001) from one year to another.The daily pattern of pollinators' visitation was related to the pattern of anthesis of the flowers of both analyzed species.
In the case of L. speciosa, the pattern of anthesis was forenoon (05:00-11:00 hrs) and the peak period of anthesis was recorded between 07:00 and 09:00 hours of the day.The stigma was receptive at the time of anthesis.The anther dehiscence started half an hour after the beginning of receptivity of the stigma.The receptivity continued for 48 hours.The amount of pollen was highest in the morning hours due to high rate of anthesis and anther dehiscence, leading to high visitation rates at that time.Highest quantitative frequency of pollinators was recorded between 07:00 to 09:00 h of the day, which also coincided with the peak timing of anthesis.
For S. wallichii, anthesis in flowers started from 06:00 h in the morning and reached to its maximum between 07:00 and 09:00 hours of the day.As the day progressed, the level of anthesis decreased and gone down to its minimum level at 11:00 h; the flower openings were again recorded during afternoon hours between 15:00 and 18:00 h, therefore, the pattern of anthesis was bimodal.Anther dehiscence started one hour after anthesis and followed a diurnal pattern.The receptivity of stigma started after 4 h of anther dehiscence and remained receptive up to 60 h.The pattern of pollinators' visitation, particularly moths, in S. wallichii was bimodal, which followed the pattern of anthesis.
Anther dehiscence offers mature pollen grains to the insects, to carry them into the receptive surfaces of stigmas, which in turn, utters the pollination success of a species and play an important part for determining the genetic make-up of populations.The detailed morphometrics of the flowers of both study species are presented in Table 1.

Pollinator visitation in Lagerstroemia speciosa
The flowers of L. speciosa were observed to be visited by 4 insect species belonging to Hymenoptera (Apis cerena, Apis florea and Xylocopa spp.) and Diptera (Chrysomya megacephala).Insects' visits to L. speciosa occurred throughout the day, but the time of maximum visitation frequency was between 07:00 and 09:00 h of day (Figs.6  and 7).Interestingly, the rate of visitation varied among individuals within the population.There was a definite interaction of floral visitors with the reproductive floral parts of L. speciosa while foraging, as shown in the Fig. 8. On the basis of visitation frequency, foraging rate and average time spent per flower, Apis cerena were the most effective pollinators, followed by Xylocopa spp., Apis florea and Chrysomya megacephala.

Discussion
Both tree study species were visited by members of bee family Apidae, with Xylocopa spp.found to be an important pollinator on the basis of its visitation frequency and rate.Flower size (large for L. speciosa and moderate for S. wallichii) along with large number of anthers per flower, copious pollen production (Khanduri et al., 2015) and morphological fitness of Xylocopa bee to the flowers (Khanduri, 2014)   species to be adapted to pollination by Xylocopa bee.Although Xylocopa bees spent less time per flower while foraging than other species, they made frequent inter-flower and inter-tree movements.This characteristic, along with its large size, ability to force into flowers, fast flight, foraging activity during strong winds (Somanathan and Borger, 2001) and long flight range of around 20 km (Kapil and Dhaliwal, 1969) are the important features for its effectiveness as a pollinator.
Apis cerena visited flowers of both study species with high frequency and thus was another important pollinator.It is one of the most widespread species of canopy pollinator along with Apis dorsata in the rainforests of Western Ghats and Srilanka (Dayanandan et al., 1990;Devy and Davidar, 2003).Apis florea was also identified as an important floral visitor of L. speciosa in the present study, as this species is known to be a key pollinator of field and orchard crops in South-Asia (Batra, 1967;Rehman et al., 1990;Sihag, 2000).However, its role in pollination of wild plants has not been documented previously (Corlett, 2004).Moreover, pollinators, especially bees and moths visitations, were significantly coincided with the pattern of anthesis in both S. wallichii and L. speciosa.Chrysomya megacephala, the Diptera belonging to the family Calliphoridae, was a significant floral visitor in terms of dispersed visitation frequency throughout the day and maximum mean time spent per flower in L. speciosa.Calliphoridae are regular visitors and important pollinators of Avecenia officinalis (a mangrove species) in South India (Aluri, 1990) and Mangifera indica in sub-tropical India (Bhatia et al., 1995).Calliphoridae are strong fliers and transmit fairly large quantities of pollen (House, 1989;Aluri, 1990), but their function as pollinators in tropical tree species are largely unexplored and underestimated.
Many members of family Lepidoptera have been reported as casual floral visitors and are considered to be less important as pollinators in lowland rainforest compared with the neotropics (Bosch, 1997;Sakai, 2001).In S. wallichii, five species of moths visited flowers irregularly, mainly during early morning and late evening hours.In the Western Ghats region of India, nine tree species have been shown to be pollinated by moths of the family Geometridae (Devy and Davidar, 2003).

Conclusions
Various moth and bee species were involved in pollination of Schima wallichii flowers.The flowers of S. wallichii showed physical fitness towards functional group of moth and bee in respect to mating success.The Xylcopa spp. of bee showed a high level of functional synchronisation with the floral morphology of L. speciosa.The smaller number of species attracted to L. speciosa means that the loss of any one or more would have a negative effect on pollination of this species.Tropical forest ecosystems sustain high degree of ecological specialisation of plant-pollinator interactions and have a highly coupled relationship between reproductive phenology and pollinators' availability.Anthropogenic pressure may decouple species interactions during pollination, which may lead to accelerated species extinction.Therefore, studies that combine observation of functional groups of pollinators with examination of pollen-mediated gene flow would greatly improve the understanding of the mechanisms involved in tree pollination.

Fig. 5 .
Fig. 5. Frequency of pollinators (bees, butterflies and beetle) visits in respect to the time of day for S. wallichii in the year 2013 (values in parentheses indicate LSD values for 5%) Fig. 6.Frequency of pollinator visits in respect to the time of day for L. speciosa in the year 2012 (values in parentheses indicate LSD values for 5%)

Table 2 .
Insects visiting the flowers of Schima wallichii in tropical forest site of Mizoram

Table 3 .
were the characteristic features of both 260 Insects visiting the flowers of Lagerstroemia speciosa in tropical forest site of Mizoram