Literature

Whilst the mammalian embryo is still in the womb, electrical activity can be detected in the developing ears (and eyes). "This spontaneous activity is required for maturation of auditory neurons and to establish auditory pathways in the brain." Recently, a big step forward has been made in understanding how this activity is generated in the ear. It involves cells in Kolliker's organ, which is a transient epithelial structure in the developing cochlea. These cells spontaneously release ATP, which goes on to activate the inner hair cells and auditory nerve fibres. It must be emphasised that this process effectively ceases at birth: "Spontaneous ATP-dependent signalling rapidly subsides after the onset of hearing, thereby preventing this experience-independent activity from interfering with accurate encoding of sound."
The research has pioneered in several ways. It has brought development into the way neuroscientists study hearing. It has identified function in an organ that was a mystery: "These [supporting cells] are cells that, until now, we thought weren't doing a hell of a lot," said Jonathan Gale, a coauthor of the study. It has shown that hearing can be stimulated by ATP: "What we were struck with was that ATP was doing the job that sound would eventually do in the developed cochlea," said Dwight Bergles of Johns Hopkins School of Medicine, corresponding author of the study. "Before the ear is mature enough to detect sound, hair cells respond to ATP." (The authors suggest that this may be the key needed to understand the onset of tinnitus). According to one commentator, the study is "essentially opening up a new area, and actually underscoring the importance of cells we normally ignore".
This appears to be a good example of how the reductionist approach to science creates blinkers. Those working in this area had a mental model of the relevant components and they were untroubled by the fact that they ignored some parts of the system. The new research shows a longitudinal dimension: cellular activity occurs in development that is relevant to maturing the system but this activity ceases and appears irrelevant in the functioning of the mature organ. However, "spontaneous activity in auditory nerve fibres before the onset of hearing is essential for the survival of target neurons in the cochlear nucleus, accurate wiring of auditory pathways, and the refinement of tonotopic maps in auditory nuclei." [emphasis added]
This is great research, moving beyond reductionism towards a holistic approach. I liked the accompanying News & Views essay which referred to a "fantasia" and to the cells of Kolliker's organ "generating their own 'virtual' music". This is a melodious sound in the biologist's ear, but it is music deriving from the beauty and complexity of the living world, not from evolutionary theory (without which, say some, nothing makes sense).
The research does have some relevance to theories of evolution. In The Edge of Evolution, Michael Behe sets out two criteria by which to judge whether random mutation hitched to natural selection is a biologically reasonable explanation for any given molecular phenomenon. These are on page 104: steps (simply the number that must be taken to reach a beneficial state) and coherence (whether those steps are random or ordered). The new role discovered for Kolliker's organ demands an extraordinarily high degree of coherence, because of its transient - but essential - contribution, whatever else may be said about steps. This points to a non-random explanation for the origin of the ear: naturally occuring mutations cannot deliver it.

The origin of spontaneous activity in the developing auditory system
Nicolas X. Tritsch, Eunyoung Yi, Jonathan E. Gale, Elisabeth Glowatzki & Dwight E. Bergles
Nature 450, 50-55 (1 November 2007) | doi:10.1038/nature06233

Abstract: Spontaneous activity in the developing auditory system is required for neuronal survival as well as the refinement and maintenance of tonotopic maps in the brain. However, the mechanisms responsible for initiating auditory nerve firing in the absence of sound have not been determined. Here we show that supporting cells in the developing rat cochlea spontaneously release ATP, which causes nearby inner hair cells to depolarize and release glutamate, triggering discrete bursts of action potentials in primary auditory neurons. This endogenous, ATP-mediated signalling synchronizes the output of neighbouring inner hair cells, which may help refine tonotopic maps in the brain. Spontaneous ATP-dependent signalling rapidly subsides after the onset of hearing, thereby preventing this experience-independent activity from interfering with accurate encoding of sound. These data indicate that supporting cells in the organ of Corti initiate electrical activity in auditory nerves before hearing, pointing to an essential role for peripheral, non-sensory cells in the development of central auditory pathways.

See also:

Forsythe, I.D., Hearing: A fantasia on Kolliker's organ, Nature 450, 43-44, (1 November 2007) | doi:10.1038/450043a

Scheff, J., Sensing through non-sensory cells, The Scientist, 31st October 2007