A flower below the tree crown
A hanging stem of Weberocereus tunilla positions its flower away from the branch on which the cactus grows. At La Selva, Costa Rica, the authors found these epiphytes high in forest crowns. Stems climbed trunks and spread along large horizontal branches. Flower-bearing portions hung below the support and left open space around the floral entrance. [Tschapka et al. 1999, pp.554–555]
The authors termed this arrangement functional flagelliflory. In many plants, flowers hang from elongated stalks. Here, their position results from development of the main cactus stem. This exposure facilitates access by a hovering animal. A possible effect of separation from foliage on echolocation detection remains an interpretation in this study. [Tschapka et al. 1999, pp.557]
Night-time rhythm and nectar
The flower opened rapidly, within approximately fifteen minutes. It had a waxy-white perianth with a dull pink tinge and a strong musky odour. It remained active for one night. Towards morning, closure began, and after 5 a.m. it became inaccessible. The observations therefore concerned nocturnal visits during a short period of floral activity. [Tschapka et al. 1999, pp.555]
Nectar secretion began after flower opening and reached its highest rate around midnight. The mean all-night quantity from twelve flowers was 354 ± 76 µl. This measurement was obtained by summing hourly collections. The figure showing the nightly pattern was based on eleven flowers. Sugar concentration rose from approximately 15% to 19% by mass and subsequently declined again. [Tschapka et al. 1999, pp.555–556, Fig.2]
A bat at the floral entrance
Bats usually fed while hovering, and an individual visit lasted less than a second. The head reached deeply into the flower. Pollen was deposited mainly on the chin and throat, from where it could be transferred to the stigma of another flower. Less often, an animal grasped the flower with its thumb claws and feet. [Tschapka et al. 1999, pp.556–557]
The study identified Glossophaga commissarisi, Hylonycteris underwoodi and Lichonycteris obscura. Cactus pollen occurred on 6.9% of 317 examined G. commissarisi and on 26.3% of 57 H. underwoodi. L. obscura was documented at a flower in a photograph. Pollen was not detected on eleven captured individuals of this rare species. Each result describes a different part of the evidence for visiting the plant. [Tschapka et al. 1999, pp.557]
Flowering and scope of evidence
Four selected plants were observed daily for a year. Flowering occurred in irregular bursts, with the greatest number of flowers between March and August. On an individual plant, several flowering days could alternate with flowerless months. Concurrent detection of pollen on bats made it possible to track floral availability beyond the four individuals included in the counts. [Tschapka et al. 1999, pp.555–557, Fig.3]
Visits, pollen placement on animals and observed fruiting support bat pollination at La Selva. The paper presents field observations without an experimental comparison of the effectiveness of individual pollinator species. Trials with other cacti in a greenhouse showed that Glossophaga soricina could visit their flowers. This separate result concerns controlled greenhouse conditions. [Tschapka et al. 1999, pp.557–558]
An account based on the complete original field-study paper.
Bat Pollination of Weberocereus tunilla, an Epiphytic Rain Forest Cactus with Functional Flagelliflory
Tschapka M., von Helversen O., Barthlott W., Bat Pollination of Weberocereus tunilla, an Epiphytic Rain Forest Cactus with Functional Flagelliflory, Plant Biology, 1999, 1(5):554-559, doi: 10.1111/j.1438-8677.1999.tb00783.x
Observations of nocturnal flowers, nectar and bats visiting Weberocereus tunilla at La Selva, Costa Rica.