Tropical Natural History 18(2): 135-145, October 2018 2018 by Chulalongkorn University

Short Note Regional Differences in Mammalian Pollinators of macrocarpa (Leguminosae): a Review

SHUN KOBAYASHI1*, TETSUO DENDA1, CHI-CHENG LIAO2, JUMLONG PLACKSANOI3, SURACHIT WAENGSOTHORN3, CHITTIMA ARYUTHAKA4, SOMSAK PANHA5,6, MASAKO IZAWA1

1Faculty of Science, University of the Ryukyus, 1 Senbaru, Nishihara, Okinawa 903-0213 JAPAN 2Department of Life Science, Chinese Culture University, 55 Huagang Road, Shilin, Taipei 11114 TAIWAN 3Sakaerat Environmental Research Station, Thailand Institute of Scientific and Technological Research, 1 Moo 9 Udom, Wang Nam Khieo, Nakhon Ratchasima 30370 THAILAND 4Faculty of Fishery, Kasetsart University, 50 Ngam Wong Wan Road, Lat Yao, Chatuchak, Bangkok 10900 THAILAND 5Department of Biology, Chulalongkorn University, Payathai Road, Pathumwan, Bangkok 10330 THAILAND 6Center of Excellence on Biodiversity, Ministry of Education and Chulalongkorn University, Bangkok 10400 THAILAND * Corresponding Author: Shun Kobayashi ([email protected]) Received: 17 May 2018; Accepted: 7 August 2018

Many angiosperms are pollinated by the challenges of pollination studies in this animals1. Among them, certain are . Mucuna macrocarpa is an evergreen pollinated exclusively by , woody vine. It produces 30–50 cm long although -pollinated are bearing 10–30 flowers with fewer in number than -pollinated purple and pale green (Fig. 1). plants2. Mammal-pollinated plants are Matured flowers emit a fermentation-like known from many taxa which reviewed by odor. Mucuna macrocarpa is distributed Fleming and Kress3. In addition, specific from Southeast Asia to Japan7. In Thailand, groups of mammals are known pollinators. it occurs in evergreen and mixed forests in The most well-known of these are , but the central to northern regions8. In the various non-flying mammals pollinate subtropics or temperate regions, it occurs in plants4. evergreen forests. The flowering season This study focuses on Mucuna changes annually and locally, and is thought (Leguminosae). Genus Mucuna includes to take place over one or two months more than 100 species and this genus is between January and May. distributed throughout tropics and Methods of Observation subtropics5. Mucuna has inflorescences with Previous observations have been either relatively large flowers and this genus has direct, in which observers stay in front of been considered to be pollinated by bats or the target flowers, or via video cameras9,10. birds specifically6. However, these methods have limitations, Among pollinators of Mucuna species, given the wariness of mammal visitors and this study reviews the behavior of Mucuna influence of observers on their flower- macrocarpa’s mammalian pollinators and visiting behavior and staying time. In its flower structure. In addition, we discuss addition, night-vision scopes are needed, TROPICAL NATURAL HISTORY. 18(2), OCTOBER 2018 136

FIGURE 1. Inflorescences of Mucuna macrocarpa. and observation of detailed flower-visiting description of flower-visiting behavior from behavior is difficult. Video cameras are photographs. expensive and battery life is short, placing a We used automatic video camera traps to limitation on observation time. In the 1990s, resolve these problems15. There are various automatic camera traps were developed for types of trigger16. Among them, cameras observation of mammalian flower visitors11. with infrared sensors which detect However, detailed observation of flower- differences between air temperature and visiting behavior is required, because not all (surface) temperature are often used flower visitors contribute to pollination, and for field surveys of mammals. In this some visitors rob nectar or drop flowers12-14, method, 1) mammals do not change their and this method does not allow for behaviors, 2) researchers can observe KOBAYASHI ET AL. – MAMMALIAN POLLINATORS OF MUCUNA MACROCARPA 137

FIGURE 2. Flower of Mucuna macrocarpa and floral structures. Banner is pale green, wing petals are purple, and carina petals are pale purple. (A): Before flower opening, (B): After flower opening, (C): Hook-like structure (before flower opening), (D): Hook-like structure (after flower opening), (E): Connected part of wing and carina petals, (F): Section of the broken line in Fig. 2A. Large amount of sweet nectar is located inside the calyx (see Fig. 2A). When the flower opens and the banner faces upward, nectar flows down from the hook- like structure (see Fig. 2D) to the tip of the carina. Dotted circle in Figs. 2C and D shows the location of the hook-like structure. throughout the day and night, and 3) flower- five petals — a banner, a pair of wings, and visiting behaviors are recorded in detail. a pair of carina petals (Fig. 2A). Nectar, a This method has been successfully used to reward for visitors, is stored inside the observe flower-visiting mammals17-19. calyx. The stamens and pistil are enclosed Explosive Opening by a pair of carina petals, preventing The flower shape of M. macrocarpa is automatic pollination. However, specific papilionaceous. Papilionaceous flowers have mammals are able to open the flower, TROPICAL NATURAL HISTORY. 18(2), OCTOBER 2018 138

FIGURE 3. Distribution map of Mucuna macrocarpa and study regions. Shaded areas show distribution. Stars indicate study sites. exposing stamens and pistil (Fig. 2B) so that allowing the explosive opener to feed21. In pollination can occur. Once the flower of M. addition, because of the wing petal’s macrocarpa is opened in this manner the attachment to the carina (Fig. 2E, F) stamens and pistil are not enclosed again. depressing the wing petal has the Additionally, the opening of the flower is simultaneous effect of pushing the carina accompanied by explosive release of pollen, down. Thus, flower opening requires the referred to as “explosive opening”. This strength of a mammal pollinator, as even mechanism is also known from other, bee- large bees, such as carpenter bees Xylocopa pollinated genera20. The flowers of these appendiculata circumvolans, cannot open plants cannot open on their own, and at least M. macrocarpa flowers22. cross-pollination by an animal, or explosive Explosive openers were considered opener, is obligatory. pollinators, but their role had not been In M. macrocarpa, flower opening takes experimentally verified9,23,24. While our place when the wing petal is pressed observations on fruit set following artificial downward, and the banner petal pollination in this genus suggest self- simultaneously pressed upward21. Thus, compatibility, fruits were not observed explosive openers must be able to when flowers did not open25. Thus, we accomplish this movement. Successful assume that explosive opening is necessary pollinators must also be able to release the for pollination in this genus, even though pair of hook-like structures at the base of the only one species was tested. banner petal, which press the wing petals Shift of Explosive Openers from both sides, preventing the flower from Explosive openers were identified M. opening automatically21 (Fig. 2C). When macrocarpa’s range, in Kyushu, Okinawa, these hook-like structures are released (Fig. and Taiwan15,18,21 (Fig. 3). Explosive 2D), nectar flows away from the calyx, openers are Japanese macaques Macaca KOBAYASHI ET AL. – MAMMALIAN POLLINATORS OF MUCUNA MACROCARPA 139

FIGURE 4. Explosive openers (effective pollinators) in three study regions. (A): Japanese macaque (Macaca fuscata), (B): ( dasymallus), (C): Red-bellied squirrel (Callosciurus erythraeus). fuscata (Fig. 4A) and Japanese martens openers are mammals. When these species Martes melampus in Kyushu; Ryukyu flying open a flower large amounts of pollen are foxes Pteropus dasymallus (Fig. 4B) in removed. Supported by the results of Okinawa; and red-bellied squirrels experiments, we conclude that explosive Callosciurus erythraeus (Fig. 4C), openers are effective pollinators. Formosan striped squirrels Tamiops In Mucuna, explosive openers comprise maritimus and masked palm civets Paguma either one or 2–3 species from the same larvata in Taiwan. Although the openers group, as reported in Table 1. Almost all of described above are available, Japanese them are bats or birds (Table 1). On the macaques are the main openers in Kyushu other hand, mammals from different orders and red-bellied squirrels are the main act as pollinators in different regions of M. openers in Taiwan, by virtue of the large macrocarpa’s range. There were other number of flowers they open when examples of plants pollinated by different compared to other openers. Explosive mammals at different sites. Traveler’s trees openers differ among regions, but all known Ravenala madagascariensis are pollinated TROPICAL NATURAL HISTORY. 18(2), OCTOBER 2018 140

TABLE 1. Explosive openers or flower visitors of Mucuna spp.

Study Explosive opener (Flower visitor**) Pollinator name Literature region/Country Species name Common name observation method M. macrocarpa Kyushu Macaca fuscata*** Japanese macaque Video camera trap / 15 (Japan) Direct observation Martes melampus Japanese marten Okinawa Pteropus dasymallus Ryukyu flying fox Direct observation 21 (Japan) Taiwan Callosciurus Red-bellied Video camera trap 18 erythraeus*** squirrel Tamiops maritimus Formosan striped squirrel Paguma larvata Masked palm civet M. sempervirens Kunming Callosciurus erythraeus Red-bellied Direct observation 49 (China) squirrel Dremomys pernyi Perny's long- nosed squirrel M. birdwoodiana* Hongkong ( leschenaulti) (Leschenault’s Direct observation 50,51 rousette ) (Paguma larvata) (Masked palm civet) M. macropoda Papua New Synycteris australis Queensland Direct observation 24 Guinea blossom bat M. gigantea* Indonesia (Bat ?) Estimate from the 23 claw mark on flower M. reticulata* Indonesia (Bat ?) Estimate from the 23 claw mark on flower M.macrophylla* Indonesia (Bat ?) Estimate from the 23 claw mark on flower M. monosperma* Indonesia (Bat ?) Estimate from the 33 claw mark on flower M. flagellipes Kamerun Woermann's fruit Direct observation 52 woermanni bat M. urens Brazil Glossophaga soricina Pallas's long- Direct observation 53 tongued bat M. holtonii Costa Rica Glossophaga Commissaris's Direct observation / 10,54 commissarisi*** long-tongued bat Video camera Hylonycteris underwoodi Underwood's long-tongued bat Lichonycteris obscura Dark long- tongued bat M. rostrata Colombia Cacicus cela Yellow-rumped Direct observation 55 cacique M. japira São Paulo Cacicus haemorrhous Red-rumped Direct observation 9 (Brazil) cacique M. sloanei* No Glossophaga soricina Pallas's long- Direct observation 56 information tongued bat M. mutisiana* No (Bat ?) No information 57**** information M. reptans* Australia (Rat or Bird ?) Estimate from the 58 floral characteristics M. pruriens* No (Bat ?) No information 57**** information *Quantitative observations were not conducted to identify the pollinator. **Parentheses indicate that explosive opening behavior was not observed for this species. *** This is a main pollinator in the region, even when multiple pollinators occur in the region. ****byArticle ruffed suggest lemursed the pollinator,Varecia but variegata we could not in find their any evidence of field observation.

KOBAYASHI ET AL. – MAMMALIAN POLLINATORS OF MUCUNA MACROCARPA 141 by ruffed lemurs Varecia variegata in their always attached to the face and does not naturally-occurring range, and by gray- touch on hand. Thus, the site of attachment headed flying foxes Pteropus poliocephalus of pollen and stigma is not always same in in Australia where traveler’s trees have been the case of macaques. introduced26,27. However, to our knowledge, Explosive opening behaviors have been pollinator shifts from one mammal to described for other Mucuna species. In M. another within a species’ natural range are japira, which is pollinated by red-rumped not known. Pollinator shifts from bats to caciques Cacicus haemorrhous, Agostini et non-flying mammals or vice versa are also al.9 noted that birds used their heads to rare. “press the base of the wing and carina petals Comparisons of Explosive Opening at the same time”. In bat-pollinated M. Behaviors macropoda, Hopkins and Hopkins24 In M. macrocarpa, it is necessary to described pollen release when the pressure simultaneously raise banner and push down (exerted by the bat with its head) on the base wing petals for explosive opening to occur. of the petals released the androecium and Does explosive opening behavior differ gynoecium from within the carina. Other among openers? Our observations clarified leguminous explosive openers are known that explosive opening behavior is common among bee-pollinated species in which the among openers, who hold wing petals with weight of bees on the plant’s wings appears their forelimbs and raise the banners with sufficient to trigger explosive opening their snouts by inserting them into the gap without any specialized behavior20,28. between wing and banner15,18,21; a single Compared to these species, the mechanism exception is the Japanese macaque, which of explosive opening in M. macrocarpa is holds the wing petal in one hand and raises more complex, its reliance on the use of the the banner petal with the other hand15. mammalian forelimb suggesting co- This difference in opening behavior may evolution with mammal pollinators. affect pollination success. The Ryukyu Flower Visitors Other than Explosive flying fox, the only opener in Okinawa, and Openers the red-bellied squirrel, the main opener in Flower visitors other than explosive Taiwan, always inserted their snouts into the openers were also recorded in the three gap between wing and banner at a similar regions sampled15,18,21. While Japanese angle18,21. This method results in both the macaques (in Kyushu) could open and adherence of pollen grains to, and contact of pollinate flowers as described, Formosan the plant’s stigma with the pollinator, on the rock macaques Macaca cyclopis could not same part of the pollinator’s body, with open flowers18. In a survey spanning at obvious consequences for pollination Nanrenshan Ecological Reserve Area, efficacy i.e. the pollinator’s method southern Taiwan16 we used camera traps maximizes the likelihood of pollen (Ltl-Acorn 5210; Shenzhen Ltl Acorn deposition on the stigma. In contrast, when Electronics Co., Ltd., China) to ascertain the Japanese macaques open flowers, pollen fate of M. macrocarpa inflorescences, grains may adhere to their hands or faces, discovering that 92.9% (n = 14) of depending on the precise opening stance inflorescences were picked and eaten by employed i.e. the distance between face and Formosan rock macaques prior to flower when opening occurs. The stigma maturation, while Formosan rock macaques TROPICAL NATURAL HISTORY. 18(2), OCTOBER 2018 142 visited mature flowers in northern Taiwan. group for studies on the diversity and Because of regional differences in evolution of mammal-dependent pollination vegetation in Taiwan29 food sources systems. Pollinators of several Mucuna exploited by macaques vary with area, and species have been clarified (Table 1). their reliance on Mucuna relative to other However, some of these pollinators were foods may also vary as a reflection of these identified only by the observations differences. Resource requirements may confirmed visits by these species but not also differ on the basis of troop size, which necessarily visits resulting in pollination varies among troops30 and may influence (Table 1). This genus has been considered to which food sources are most regularly be pollinated by bats or birds6, but this visited. suggestion is not based on enough surveys. Secondly, visitors to opened flowers For example, although van der Pijl23,33 were observed in all regions. One of them, conducted pioneering studies, he concluded the honeybee, Apis cerana visits to collect that bats were pollinators based on the claw pollen, sometimes coming into contact with marks on flowers. While claw marks of fruit the stigma15,18 and potentially acting as a bats may differ from those of other secondary pollinator, although efficiency is mammals, squirrels and macaques also mark unknown. In addition, honeybees frequently flowers (Fig. 5) and it is sometimes difficult visited flowers immediately after explosive to distinguish between the marks of bats and opening. Diurnal Japanese macaques and those of other mammals. In addition, red-bellied squirrels frequently visited information on flower-visiting frequency flowers in the early morning25, thus and flower-visiting behavior is lacking, and attracting the visits of honeybees at this van der Pijl23 did not consider other possible time. visitors. Furthermore, the genus Mucuna Interestingly, birds seldom visited M. requires specific flower-visiting behaviors macrocarpa flowers. Japanese white-eyes (this requirement does not apply only to M. Zosterops japonicus and brown-eared macrocarpa). Thus, detailed observations bulbuls Hypsipetes amaurotis visited are needed especially in those species. flowers less frequently than explosive Otherwise, Mucuna species may have other openers both in Kyushu and Okinawa, and additional pollinators, and pollinators may did not open flowers at all15,18,21. In Taiwan, differ in different parts of Mucuna’s range. no birds visited flowers18. These results The genus Mucuna originated in tropical suggest that M. macrocarpa relies solely on Asia34. Therefore, in order to understand the mammal pollinators. While the Okinawa pollination system of Mucuna, Southeast woodpecker Sapheopipo noguchii visited Asia is the most important region. The fauna relatively frequently and fed on nectar31 this in the tropical zone is more diversified than may reflect a feeding habit related to that in the temperate zone, and squirrels and evolution on a small island32, and carnivores are the most diversified group in insignificant in terms of pollination. Southeast Asia35. The importance of flying Future Challenges foxes36-38 and squirrels38,39 as pollinators in It will be instructive to identify the Southeast Asia has been noted. In fact, pollinators of Mucuna spp. A large number diurnal tree squirrels act as pollinators of of vertebrate-pollinated plant species occur number of plants besides Mucuna40-42. Most in the tropics3, and Mucuna is a good target studies of non-flying mammal-pollinated KOBAYASHI ET AL. – MAMMALIAN POLLINATORS OF MUCUNA MACROCARPA 143

FIGURE 5. Claw marks of explosive openers. (A) and (B): Ryukyu flying-fox, (C): Japanese macaque, (D): Red- bellied squirrel. Arrows indicate the claw marks of each opener. plants have been conducted in Australia, and common floral traits such as shape, nectar South Africa, and more research into secretion dynamics, and odor (volatile mammal-dependent pollination systems in components). Although several Mucuna Southeast Asia are needed. species fit bat-pollination syndromes10,24,43, One of the important issues for floral traits are seldom quantitatively discussion of the evolution of mammal- assessed for their fit with pollinators. This pollinated plants is that of floral traits. Many represents a significant gap in our plants that are considered bat-pollinated knowledge, and there are instances in which species have floral traits reflecting this pollinators predicted for a species, based on pollination syndrome, causing researchers to floral characteristics and pollination overlook the possibility of other pollinators. syndromes, differ from observed pollinators. Pollination syndrome is the concept that Many such exceptions have recently been plants pollinated by same pollinator have described44-46 and it appears that prediction TROPICAL NATURAL HISTORY. 18(2), OCTOBER 2018 144 of pollinators based on the pollination needed to enhance understanding of the syndrome is unrealistic. However, it is evolution of mammal pollination. known that a specific floral trait attracts a specific pollinator. In mammal-pollinated ACKNOWLEDGEMENTS plants, volatile components are one of the most important attractive traits47. Further We appreciate to T. Doi, T. Iwamoto, S. research on Mucuna is needed to clarify Mashiba, J. Kawano, H. Ui, Y.-H. Lin, K.- floral traits such as shape, nectar production F. Lin, and S.-H. Wu for supporting field dynamics, and volatile components, and to surveys. This study was partly supported by examine the relationship of these traits with JSPS KAKENHI (Grant number pollinators. 16H05771). Conclusion

Pollinators of M. macrocarpa differ among regions but all of them initiate LITERATURE CITED explosive opening, with the exception of the Japanese macaque. Such commonness of 1. Ollerton, J., Winfree, R. and Tarrant, S. 2011. How many flowering plants are pollinated by behavior indicates that the ability of M. ?. Oikos 120:321-326. macrocarpa to open explosively makes it 2. Devy, M.S. and Davidar, P. 2003. Pollination available to a number of potential system of trees in Kakachi, a mid-elevation wet pollinators. However, in all regions, non- evergreen forest in Western Ghats, India. American Journal of Botany 90:650-657. mammal flower visitors are infrequently 3. Fleming, T.H. and Kress, W.J. 2013. The observed, suggesting a reliance on mammals Ornament of Life: Coevolution and at the expense of other groups, and the Conservation in the Tropics. The University of evolution of flower strength useful for Chicago Press, Chicago. 588 pages. prohibiting small animals from robbing 4. Carthew, S.M. and Goldingay, R.L. 1997. Non- flying mammals as pollinators. Trends in nectar. Ecology and Evolution 12:104-108. The explosive opening behavior of 5. Schrire, B.D. 2005. Tribe . In: Lewis, Japanese macaques is distinctive, and this G., Schrire, B.D., Mackinder, B. and Lock, M. pollinator occurs at the northern limit, or (eds.). Legumes of the World. Royal Botanic 15 Gardens, Kew, London, UK. pp. 393-431. edge, of M. macrocarpa’s range , where 6. Fægri, K. and van der Pijl, L. 1979. The 48 pollinator shift is more likely to occur . In Principles of Pollination Ecology, 3rd edition. addition, Japanese macaques dropped Pergamon Press, Oxford, UK. 256 pages. flowers frequently15 and other Macaca 7. Tateishi, Y. and Ohashi, H. 1981. Eastern monkeys cannot open and pollinate flowers. Asiatic species of Mucuna (Leguminosae). The Botanical Magazine, Tokyo 94:91-105. Thus, Japanese macaques might have 8. Wilmot-Dear, C.M. 2008. Mucuna Adans. become pollinators as a secondary (Leguminosae) in Thailand. Thai Forest Bulletin consequence of learning to prey on Mucuna (Botany) 36:114-139. flowers. 9. Agostini, K., Sazima, M. and Sazima, I. 2006. Bird pollination of explosive flowers while In Southeast Asia, non-flying mammals foraging for nectar and caterpillars. Biotropica are known to pollinate plants reported as 38:674-678. bat-pollinated species18,44. Thus, we need to 10. von Helversen, D. and von Helversen, O. 2003. clarify the pollinators of other Mucuna Object recognition by echolocation: a nectar- species in Asia. In addition, clarifying feeding bat exploiting the flowers of a rain forest vine. Journal of Comparative Physiology pollination systems, including floral traits, is KOBAYASHI ET AL. – MAMMALIAN POLLINATORS OF MUCUNA MACROCARPA 145

A 189:327-336. M. 2018. Who can open the flower? Assessment 11. Carthew, S.M. 1991. Monitoring animal activity of the flower opening force of mammal- with automated photography. Journal of pollinated Mucuna macrocarpa. Plant Species Wildlife Management 55:689-692. Biology. DOI: 10.1111/1442-1984.12221 12. Grand, T.I. 1972. A mechanical interpretation of 23. van der Pijl, L. 1941. Flagelliflory and terminal branch feeding. Journal of Mammalogy cauliflory as adaptations to bats in Mucuna and 53:198-201. other plants. Annals of the Botanic Gardens, 13. Janson, C.H., Terborgh, J. and Emmons, L.H. Buitenzorg 51:83-93. 1981. Non-flying mammals as pollinating 24. Hopkins, H.C.F. and Hopkins, M.J.G. 1993. agents in the Amazonian forest. Biotropica 13:1- Rediscovery of Mucuna macropoda 6. (Leguminosae: Papilionoideae), and its 14. Deng, X.-B., Ren, P.-Y., Gao, J.-Y. and Li, Q.-J. pollination by bats in Papua New Guinea. Kew 2004. The striped squirrel (Tamiops swinhoei Bulletin 48:297-305. hainanus) as a nectar robber of ginger (Alpinia 25. Kobayashi, S. 2017. Mammal-dependent kwangsiensis). Biotropica 36:633-636. pollination system of Mucuna macrocarpa 15. Kobayashi, S., Denda, T., Mashiba, S., (). Ph.D. Thesis, University of the Iwamoto, T., Doi, T. and Izawa, M. 2015. Ryukyus, Okinawa, Japan. 139 pages. Pollination partners of Mucuna macrocarpa 26. Calley, M., Braithwaite, R.W. and Ladd, P.G. (Fabaceae) at the northern limit of its range. 1993. Reproductive biology of Ravenala Plant Species Biology 30:272-278. madagascariensis Gmel. as an alien species. 16. Swann, D.E., Kawanishi, K. and Palmer, J. Biotropica 25:61-72. 2011. Evaluating types and features of camera 27. Kress, W.J., Schatz, G.E., Andrianifahanana, M. traps in ecological studies: a guide for and Morland, H.S. 1994. Pollination of researchers. In: O’Connell, A.F., Nichols, J.D. Ravenala madagascariensis (Strelitziaceae) by and Karanth, K.U. (eds.). Camera Traps in lemurs in Madagascar: evidence for an archaic Animal Ecology, Methods and Analyses. coevolutionary system? American Journal of Springer Japan, Tokyo, Japan. pp. 27-43. Botany 81:542-551. 17. Hobbhahn, N., Steenhuisen, S.L., Olsen, T., 28. Kishi, S. 2015. Flower-visiting behavior of Midgley, J.J. and Johnson, S.D. 2017. native and alien Xylocopa carpenter bees to Pollination and breeding system of the Japanese Wisteria flowers. Japanese Journal of enigmatic South African parasitic plant Entomology (New Series) 18:31-38. In Japanese Mystropetalon thomii (Mystropetalaceae): with English summary. rodents welcome, but not needed. Plant Biology 29. Liao, C.-C., Liu, M., Su, M.-H. and Wang, J.-C. 19:775-786. 2014. Compression and overlap of unique 18. Kobayashi, S., Denda, T., Liao, C.-C., Wu, S.- vegetation system of subtropical mountain H., Lin, Y.-H. and Izawa, M. 2017. Squirrel resembling tropical and temperate forests along pollination of Mucuna macrocarpa (Fabaceae) elevation. Journal of Forest Research 19:215- in Taiwan. Journal of Mammalogy 98:533-541. 225. 19. Kühn, N., Midgley, J. and Steenhuisen, S.-L. 30. Hsu, M. and Lin, J.-F. 2001. Troop size and 2017. Reproductive biology of three co- structure in free-ranging Formosan macaques occurring, primarily small-mammal pollinated (Macaca cyclopis) at Mt. Longevity, Taiwan. Protea species (Proteaceae). South African Zoological Studies 40:49-60. Journal of Botany 113:337-345. 31. Kobayashi, S., Denda, T. and Izawa, M. 2014. 20. Westerkamp, C. 1997. Keel blossoms: Bee Okinawa Woodpeckers Sapheopipo noguchii flowers with adaptations against bees. Flora (Picidae) robbing Mucuna macrocarpa. Strix 192:125-132. 30:135-140. In Japanese with English summary. 21. Toyama, C., Kobayashi, S., Denda, T., 32. Kotaka, N., Ozaki, K., Toguchi, Y., Kinjo, M. Nakamoto, A. and Izawa, M. 2012. Feeding and Ishida, K. 2006. Extraordinary sexual behavior of the Orii’s flying-fox, Pteropus differences in foraging niche in the Okinawa dasymallus inopinatus, on Mucuna macrocarpa woodpecker on a subtropical island. Journal of and related explosive opening of petals, on Ornithology 147:196. Okinawajima Island in the Ryukyu Archipelago, 33. van der Pijl, L. 1956. Remarks on pollination by Japan. Mammal Study 37:205-212. bats in the genera Freycinetia, Duabanga, and 22. Kobayashi, S., Hirose, E., Denda, T. and Izawa, Haplophragma, and on chiropterology in general. TROPICAL NATURAL HISTORY. 18(2), OCTOBER 2018 146

Acta Botanische Neerlandica 5:135-144. 45. Ollerton, J., Alarcōn, R., Waser, N.M., Price, 34. Moura, T.M., Vatanparast, M., Tozzi, M.V., Watts, S., Cranmer, L., Hingston, A., A.M.G.A., Forest, F., Wilmot-Dear, C.M., Peter, C.I. and Rotenberry, J. 2009. A global test Simon, M.F., Mansano, V.F., Kajita, T. and of the pollination syndrome hypothesis. Annals Lewis, G.P. 2016. A molecular phylogeny and of Botany 103:1471-1480. new infrageneric classification of Mucuna 46. Amorim, F.W., Galetto, L. and Sazima, M. Adans. (Leguminosae-Papilionoideae) including 2013. Beyond the pollination syndrome: nectar insights from morphology and hypotheses about ecology and the role of diurnal and nocturnal biogeography. International Journal of Plant pollinators in the reproductive success of Inga Sciences 177:76-89. sessilis (Fabaceae). Plant Biology 15:317-327. 35. Corlett, R.T. 2007. What’s so special about 47. Johnson, S.D., Burgoyne, P.M., Harder, L.D. Asian tropical forests?. Current Science and Dötterl, S. 2011. Mammal pollinators lured 93:1551-1557. by the scent of a parasitic plant. Proceedings of 36. Marshall, A.G. 1983. Bats, flowers and fruit: the Royal Society B 278:2303-2310. evolutionary relationships in the Old World. 48. Johnson, S.D. 2010. The pollination niche and Biological Journal of the Linnean Society its role in the diversification and maintenance of 20:115-135. the southern African flora. Philosophical 37. Cox, P.A., Elmqvist, T., Pierson, E.D. and Transactions of the Royal Society of London, Rainey, W.E. 1991. Flying foxes as strong Series B 365:499-516. interactors in south pacific island ecosystems: A 49. Chen, G., Zhang, R., Dong, K., Gong, W. and conservation hypothesis. Conservation Biology Ma, Y. 2012. Scented nectar of Mucuna 5:448-454. sempervirens and its ecological function. Sheng 38. Corlett, R.T. 2004. Flower visitors and Wu Duo Yang Xing 20:360-367. In Chinese pollination in the Oriental (Indomalayan) with English summary. region. Biological Review 79:497-532. 50. Lau, M. 2004. Bat pollination in the climber 39. van der Pijl, L. 1961. Ecological aspects of Mucuna birdwoodiana. Porcupine! 30:11-12. flower evolution. II. Zoophilous flower classes. 51. Lau, M.W.N. 2012. Masked palm civet Paguma Evolution 15:44-59. larvata apparently feeding on nectar of Mucuna 40. McCann, C. 1934. The flying fox (P. giganteus) birdwoodiana. Small Carnivore Conservation and the palm squirrel (F. tristiatus) as agents of 47:79-81. pollinization in (Grevillea robusta A. Cunn.) the 52. Grünmeier, R. 1993. Bestäubung der Fabaceae silky oak. Journal of Bombay Natural History Mucuna flagellipes durch flughunde in Society 36:761-764. Kamerun. In: Barthlott, W., Naumann, C.M., 41. Yumoto, T., Momose, K. and Nagamasu, H. Schmidt-Loske, K. and Schuchmann, K.-L. 2000. A new pollination syndrome–squirrel (eds.). Animal-Plant Interactions in Tropical pollination in a tropical rain forest in Lambir Environments. Museum A. Koenig, Bonn, Hills National Park, Sarawak, Malaysia. Tropics Germany. pp. 29-39. In Germany. 9:147-151. 53. Sazima, I. and Sazima, M. 1978. Polinização 42. Tandon, R., Shivanna, K.R. and Mohanram, por morcegos em (Leguminosae). H.Y. 2003. Reproductive biology of Butea In: Resumos da XXX Reunião da Sociedade monosperma (Fabaceae). Annals of Botany Brasileira para o Progresso da Ciência. p. 419. 92:715-723. 54. Baker, H.G. 1970. Two cases of bat pollination 43. Agostini, K., Sazima, M. and Galetto, L. 2011. in Central America. Revista de Biología Nectar production dynamics and sugar Tropical 17:187-197. composition in two Mucuna species 55. Cotton, P.A. 2001. The behavior and (Leguminosae, ) with different interactions of birds visiting Erythrina fusca specialized pollinators. Naturwissenschaften flowers in the Colombian Amazon. Biotropica 98:933-942. 33:662-669. 44. Rivera-Marchand, B. and Ackerman, J.D. 2006. 56. Zies, P. 1999. Batty about sea-. The Bat pollination breakdown in the Caribbean Drifting 5:6-7. columnar cactus Pilosocereus royenii. 57. Stroo, A. 2000. Pollen morphological evolution Biotropica 38:635-642. in bat pollinated plants. Plant Systematic and Evolution 222:225-242.