Not much sunlight penetrates deep in the ocean, but this does not mean that the depths are completely dark. Julien Claes from the Catholic University of Louvain, Belgium, explains that many deep ocean creatures produce their own light. But while the light production mechanisms of bony fish are quite well understood, almost nothing was known about the way that luminescent elasmobranchs produce light. `Luminescent sharks live in the deep sea and you need live animals to study luminescence, but it is difficult to keep them alive at the surface,' explains Claes. However, when Claes came across a paper by a muscle physiologist, Harald Kryvi, describing how he had kept velvet belly lantern sharks alive in a Norwegian fjord, he realised that this could be the breakthrough. In winter, the surface temperature in Norwegian Fjords is the same as that at depth, so the fish survive the ascent. Claes and his supervisor, Jerome Mallefet, headed north to the Raunefjord to catch live velvet belly lantern sharks and test how they regulate their bioluminescence(p. 3684).

Laying long lines on the fjord bottom, Claes and Mallefet routinely caught 30–40 velvet belly sharks on their fishing trips before returning to Bergen where they kept the fish in tanks. According to Claes, the shark's light producing organs (photophores) are much tinier (<150 μm diameter)than bony fish's photophores (up to 1 cm diameter), and the shark can have as many as 2000 cm–2 distributed across their bellies and fins. Dissecting 0.55 cm diameter patches of skin from the fish's belly, Claes and Mallefet injected neurotransmitters, such as adrenaline and GABA, into the skin and measured the light produced with a luminometer, to test whether the photophores are controlled by the shark's nervous system, but were unable to stimulate the skin to glow.

Having ruled out neural control, Claes and Mallefet turned to three hormones that are known to regulate skin coloration in sharks: melatonin,prolactin and α-MSH. Injecting melatonin into a skin patch and shutting it inside the luminometer, Claes and Mallefet were amazed to see a perfect luminosity curve plotted on the computer screen. `It was a fantastic moment,'says Claes, `just amazing'. And when the duo repeated the experiment with prolactin, the skin glowed again. Both hormones stimulated the skin to glow,with melatonin producing a long weak glow, while prolactin generated a shorter brighter burst of light. And when the duo applied α-MSH to a piece of skin before stimulating it with melatonin, the skin would not glow.α-MSH inactivated the photophores. Claes and Mallefet had found the skin's on and off switches.

But why use slow acting hormones to activate bioluminescence when bony fish use fast acting nerves? Claes explains that sharks probably use bioluminescence for two reasons: camouflage against background light from the surface and communication. As shark melatonin levels depend on light levels in the environment, this could be a perfect mechanism allowing sharks to match their own luminosity with the background – and remain invisible to predators and prey hovering below – while prolactin activates the glowing photophores briefly and rapidly, probably for communication with members of their own species.

Claes, J. M. and Mallefet, J. (
). Hormonal control of luminescence from lantern shark (Etmopterus spinax)photophores.
J. Exp. Biol.