Colour is something we take for granted. But while it is useful during the day, in dim light our colour vision fails because of the low levels reaching our eyes' colour receptors. It was generally assumed, therefore, that nocturnal creatures would not have colour vision, so it was a surprise when Almut Kelber and colleagues discovered that it exists in a nocturnal hawkmoth. Hawkmoths forage for nectar using their colour vision to find the best flowers. Most hawkmoths, whether diurnal or nocturnal, are active at dawn and dusk, but at this time of day the light levels change drastically. This makes finding the right coloured flower difficult, but both types of insect seem to be up to the challenge. This ability to constantly recognise colour regardless of the colour of the illumination is called colour constancy. If we look at a blue flower on a sunny day, our eyes register a specific spectrum of light that is reflected back from the flower. But if a cloud passes over the sun,the surrounding light changes; `if you look at the spectrum that reaches the eyes, it is very different,' explains Anna Balkenius from Lund University in Sweden. Despite this we still know that the flower is blue. Colour constancy has only been found in a few species so far, but all have colour vision, and it would clearly be an advantage for hawkmoths. But whether a nocturnal hawkmoth would still have this ability was a question that Balkenius set about testing with Kelber (p. 3307). `Colour constancy appealed to me because it has a philosophical dimension and isn't easy to understand. I thought it would be fun to test it in hawkmoths,' enthuses Balkenius.

First, the team trained moths to associate a sugary treat with a specifically coloured paper `flower' under white light. Balkenius then removed the tasty reward and tested the moths' colour preferences by offering them the choice of identical colours viewed under different coloured lights to see which colour they chose. Even though one of the test colours had a similar spectrum under coloured light to the rewarded colour under white light, the moths seemed able to discriminate between the colours, and preferred visiting the coloured flower that had provided food during training.

But Balkenius needed convincing that the moths hadn't chosen this colour for other reasons; shuffling the flowers prevented the hawkmoths from associating the position of a flower with food, but this did not eliminate the possibility that they were learning which colour provided food relative to the other colours available. So, in another experiment, Balkenius trained the moths to choose the bluest of two colours. Then they were tested with two colours, one the previously rewarded colour and the other even bluer. `I tried to trick them,' laughs Balkenius. But the hawkmoths were not so easily fooled and went for the colour that they associated with food.

So what's next? Balkenius has some big ideas - relative colour learning has been studied in very few animals and she would like to investigate if dogs use it, especially as they have a more limited range of colour vision. But for now she is sticking with her hawkmoths and is currently investigating whether they use other senses alongside vision to guide them to their dinner.

Balkenius, A. and Kelber, A. (
). Colour constancy in diurnal and nocturnal hawkmoths.
J. Exp. Biol.