ObjectiveThe concept of pollination syndromes provides a classic hypothesis for the evolution of plants and pollinators. However, in recent years, with more and more generalized pollination systems reported, pollination syndromes have been questioned gradually. Cymbidium qiubeiense in the Subgen. Jensoa, is a terrestial orchid with fragrant flower and purplish spots on the labellum. These morphological features are similar to other orchids of Subgen. Jensoa, such as C. lancifolium and C. goeringii, which are pollinated by Apis cerana cerana. According to the theory of pollination syndromes, we predicted that C. qiubeiense was also pollinated by A. cerana cerana. The purpose of our study is to verify the accuracy of pollination syndromes in predicting the pollinators, and explore its scope of application.

MethodPollination biology of C. qiubeiense was investigated during 2010-2012 in the Yachang Orchid National Natural Reserve, which is located in Guangxi of southern China. We investigated its flowering phenology, flower form, pollination behavior of insect, flower color, flower odor and breeding system.

ResultThe unique pollinator of C. qiubeiense was A. cerana cerana, and no rewards could be found in the flowers. It was speculated that C. qiubeiense had food-deceptive pollination mechanism and C. qiubeiense attracted A. cerana cerana by the purplish spots on the labellum (false nectar). Phenols, aldehydes and esters were identified in its odor, and aldehydes were considered to play an important role in attracting pollinators. We successfully verified the accuracy of pollination syndromes in predicting pollinators.

ConclusionThis study provides a case for the scientific assumptions that the pollination syndromes can predict pollinators accurately. However, the pollination syndromes are composed of a suite of dynamic and evolutionary floral trait. It is necessary to take habitat change, secondary pollinators, pollinator function group, history evolution and gene into account when we use pollination syndromes to predict the pollinators on the premise of knowing the pollinators of closely related plants.