Why do flowers have such beautiful colours? The quick answer that you’ll probably think of is, “To attract pollinators.”
This New Zealand bluebell (Wahlenbergia albomarginata) is – despite the name – usually mostly white. There is variation in this species, though; you can see some of this in the picture at right. Most of the related species are blue, and the typical explanation for why this species is white is that the pollinators that visit the flower dislike blue.
Campbell and colleagues argue that the hypothesis of pollinator preference for colours is often not very well tested. This surprised me since, like you, I’d hear the story about flowers being brightly coloured to attract pollinators so often. They set out with a very thorough series of experiments to test the relationship of pollination to colour.
The researchers’ first step was simply to observe. No experimental manipulations of any sort; they just watched pollinators visit flowers. They found no difference in visits to flowers based on colour, though they saw a preference for size.
Then they busted out experimental manipulations. They painted the flowers. The native bees showed no preference for white or blue, speaking against the hypothesis. Flies disliked the blue, which was consistent with the original hypothesis.
But... about 90% of pollen transfer (depending on flower colour) was by bees (who didn’t mind the blue) rather than flies. And pollen export (as estimated with dyes) correlated highly with bee visitation, but not with fly visitation.
The team didn’t find any difference in moving pollen from plant to plant that was associated with the flowers’ natural colour.
Finally, they did some larger scale experiments. The results were a bit complicated. The pollinators’ choices changed, depended on how much variation in colour the pollinators got to see, and on how big the flower beds were.
When the pollinators had a natural range of colours, they still showed no preference.
When the experimenters enhanced the colours by painting some bluebells even brighter blue, the pollinators show no preference in a small plot of flowers, They finally showed a preference for a flower colour in a larger plot of flowers!
Except it was in the the wrong direction: a preference for blue, not white.
All in all, you’d expect there to be a lot more blue flowers in this species.
The authors explored some alternative hypotheses that were unrelated to pollination. One was a possibility that there was some sort of thermal advantage to differences in colour, but there were no temperature differences between the blues and whites.
Another possibility was that there was some selective pressure from herbivores, but the team found (say it with me) no differences in how much the plant-eaters were chomping on the blues and whites.
Having eliminated a lot of functional explanations related to the bluebells’ ecology, the authors suggest that the flower colour is incidental, and there's some other factor under selection that is linked to flower colour by genetic happenstance. They’re betting on proteins, anthocyanins, that are related to both colour and to temperatures stress.
Campbell and the rest are busy breeding a bright blue and a white strain for new experiments and comparisons. Watch this space.
Campbell D, Bischoff M, Lord J, Robertson A. 2011. Where have all the blue flowers gone: pollinator responses and selection on flower colour in New Zealand Wahlenbergia albomarginata. Journal of Evolutionary Biology: in press. DOI: 10.1111/j.1420-9101.2011.02430.x
Photo by Mollivan Jon on Flickr; used under a Creative Commons license.