Anyone lucky enough to dive on a coral reef knows they are truly breath taking. And the colours must be even more vibrant to the species that live there. Or are they? Ulrike Siebeck explains that although many marine species have several photoreceptors that detect light of different wavelengths, there was no direct evidence that they actually perceive colour; `you need behavioural experiments to demonstrate that fish are using it' says Siebeck. So she teamed up with Guy Wallis and Lenore Litherland and headed off to Lizard Island Research Station on the Great Barrier Reef to test out Ambon damselfish's colour vision (p. 354).

According to Siebeck, capturing the yellow fish was relatively straightforward. Equipped with a hand net and Ziploc bag, she and Litherland went SCUBA diving, trapping fish on the island's reefs ready to test their colour recognition skills. But learning how to train the fish was far more tricky; the team had to get into `fish psychology' to learn how to tell the fish what to do. Fortunately the fish turned out to be quick learners,`possibly because they are territorial and quickly recognise novel objects placed in their territory' explains Siebeck. According to Siebeck the damselfish try to nudge intruders out of their territory. So she and Wallis took advantage of this behaviour and trained each fish to nudge 10 times at a coloured latex finger before rewarding them with a fish food snack. Having trained one group of fish to recognise a rubber finger painted yellow and another group to recognise a finger painted blue, Siebeck and Wallis offered each fish a choice between blue and yellow fingers and watched to see which colour the fish opted for. Amazingly the yellow trained fish selected the yellow finger on 95% of occasions, and the blue trained fish got the blue finger more than 91% of the time.

But were the fish simply differentiating between light and dark colours, or genuinely distinguishing between blue and yellow? Siebeck and Wallis added black or white paint to the yellow and blue paints to darken or lighten the colours before testing whether the fish could distinguish the different shades. Offering the fish a choice between their trained colour and one of three shades of the other, the fish correctly nudged at their trained colour over 90% of the time. And when the team offered the fish a choice between a lighter or darker shade of their trained colour and the distractor colour,they successfully selected the colour they'd been trained to recognise almost 90% of occasions. Finally Siebeck offered the fish the choice between a shade of their trained colour and a shade of the distractor colour. Again the Damselfish consistently recognised their trained colour, regardless of brightness: they have colour vision. `We were very excited that they could do it so well' says Siebeck.

Having found that the fish are remarkably quick learners, and that their colour vision is surprisingly accurate, Siebeck is confident that the damselfish can teach us more about how they see their vibrant watery world.

Siebeck, U. E., Wallis, G. M. and Litherland, L.(
). Colour vision in coral reef fish.
J. Exp. Biol.