Mauritian red nectar remains a mystery


Floral nectar is rich in chemicals and induces pollination1. Although it may be tainted by algae or mould, it usually lacks colouring agents. However, a few plant species in Mauritius break this rule and produce red nectar. We attempted to find a function for this coloration, but its role remains unclear.


In 1976, a new species of plant in the family Campanulaceae, Nesocodon mauritianus (Fig. 1a), was found in Mauritius growing on inaccessible cliffs near the huge waterfall Cascade Cinq Cents Pied2. This is the only known population and in 1997 we counted just 110-130 plants. So N. mauritianus is among the many endangered Mauritian species3. It was noted that the species' floral nectar was scarlet-red4 (Fig. 1a), an unusual characteristic that was subsequently overlooked.

Figure 1: Red-coloured nectar in the Mauritian flower Nesocodon mauritianus.

a, Red nectar flowing from a flower of N. mauritianus. b, Structure of the red-colouring aurone. c, N. mauritianus visited by a red-whiskered bulbul (Pycnonotus jocosus).

We observed two bird species visiting flowers of N. mauritianus: the introduced red-whiskered bulbul (Pycnonotus jocosus) (Fig. 1c) and the native merle (Hypsipetes olivaceus). The bulbul pierced the base of the flower and robbed the nectar (80.8% of 52 flower visits, 40 hours of observation of 5 flowering plants), inserted its head into the flower to reach the nectar (9.6%) or tore the flower into pieces (9.6%). Only one merle was observed, and it picked an entire flower and flew away with it. No other animal was seen near the plants.

We mass-propagated N. mauritianus through seeds and analysed its nectar. We found that the nectar's pH was as high as 9.2 (the known pH range of all species is 3-10;1). When placed in acid, the red nectar turned yellow and showed antibacterial activity.

Using proton nuclear magnetic resonance spectroscopy, we identified the red pigment as a temperature-sensitive aurone (Fig. 1b). The yellow pigment was a light-sensitive 3-glucosylated flavone, or a flavonol with the same B-ring substitutions as the aurone. The corolla pigment was an anthocyanine with delphinidine as aglycon. All the pigments were products of flavonoid biosynthesis.

The nectar also contained 22-25% w/w hexose-dominated sugar. No specific floral scent was noticed, but hexanal, 2-hexenal, hexenol and hexanol were identified from leaves using gas-chromatography-mass spectrometry.

We examined other Mauritian flowers for red nectar and found it in Trochetia blackburniana (a species belonging to the family Sterculiaceae). In addition, T. boutoniana was reported locally5 to produce red nectar, and this was confirmed. We measured the sugar content of the latter as 25-30% and found that its red pigment was an aurone too. Trochetias are pollinated and nectar-robbed by native white-eye birds (Zosterops borbonicus and Z. chloronothos)5,6,7.

These three bird-pollinated Mauritian endemics are the only known plants in the world with coloured nectar. The red nectar in the blue flower of N. mauritianus may improve attractiveness to birds, but why do the red-flowered Trochetias produce red nectar?

We envisage at least three explanations for the evolution of this unique character: the original pollinator was an endemic but now extinct member of the Mauritian “afterlife”8 (recently extinct species); the red colour is an honest signal9 to pollinators improving their foraging efficiency and generating a fitness advantage over other bird-pollinated species; or the red pigment is associated with a deterrent against nectar robbers.

We regard the last two possibilities as unlikely. If red nectar is an honest signal, then why has it failed to evolve in other regions and taxa in which bird-pollination occurs? If it is a warning, then it is a poor ‘scarecrow’, as these taxa are heavily robbed of nectar.

We will search for extinct pollinators among recently extinct Mauritian species. However, “ghosts are not easy to pin down”8.


  1. 1

    Baker, H. G. & Baker, I. in Coevolution of Animals and Plants (eds Gilbert, L. E. & Raven, P. H.) 100–140 (Univ. Texas Press, Austin, 1975).

  2. 2

    Richardson, I. B. K. Kew Bulletin 33, 547–550 (1979).

    Google Scholar 

  3. 3

    Wyse Jackson, P. S., Strahm, W., Cronk, Q. C. B. & Parnell, J. A. N. Moorea 7, 35–45 (1988).

    Google Scholar 

  4. 4

    Wyse Jackson, P. S. Kew Magazine 7, 113–117 (1990).

    Google Scholar 

  5. 5

    Staub, F. Proc. R. Soc. Arts Sci. Mauritius 5, 7–78 (1987).

    Google Scholar 

  6. 6

    Safford, R. J. Dodo. J. Jersey Wildlife Preserv. Trust 27, 113–138 (1987).

    Google Scholar 

  7. 7

    Staub, F. Fauna of Mauritius and Associated Flora (Indian Ocean, Rose-Hill, 1993).

  8. 8

    Lawton, J. H. Oikos 73, 145–147 (1995).

    Google Scholar 

  9. 9

    Zahavi, A. J. Theor. Biol. 53, 205–214 (1975).

    Google Scholar 

Download references

Author information



Rights and permissions

Reprints and Permissions

About this article

Cite this article

Olesen, J., Rønsted, N., Tolderlund, U. et al. Mauritian red nectar remains a mystery. Nature 393, 529 (1998).

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.