Variability of interspike intervals in optic nerve fibers of Limulus: effect of light and dark adaptation.

The intervals between the impulses discharged by afferent neurons under steady conditions usually vary considerably. This variability is of interest for several reasons: (1) It must depend in some way on the underlying receptor and neural mechanisms that generate and propagate the impulses. An analysis of the factors that influence variability can therefore be expected to yield some indication of the nature of those mechanisms. (2) The information transmitted to the central nervous system by afferent neurons is coded in terms of the intervals between impulses, that is, in terms of the temporal pattern of the discharge rather than in terms of the shapes or amplitudes of the individual impulses. Any intrinsic variability in the intervals between impulses discharged by the neuron must therefore limit its capacity to carry information about extrinsic events. (3) Although the intrinsic variability may be "noise" as far as external events are concerned, there is nevertheless the possibility that it may actually carry useful information to the central nervous system about the state of the receptor or neuron and the influences that contribute to the variability of the discharge. The variation of the intervals between impulses has been investigated and described for many different types of neurons,'-3 but the causes of the variability are largely unknown. One supposed cause is "biological noise"--minute haphazard fluctuations in membrane potential such as those first observed by Fatt and Katz4 at motor nerve endings in muscle. Recently, for example, it was shown that such random fluctuations, observed in spinal motoneurons of the cat, are adequate to account for the variability of the intervals between impulses discharged by these neurons.5 It is usually difficult to control the random fluctuations or "noise" in the membrane potential of a discharging neuron. But such fluctuations are unlikely to be altogether haphazard, and with adequate knowledge of factors that influence their frequency, amplitude, and other characteristics, they may sometimes be brought under control. This is the case with the irregular fluctuations in membrane potential observed by Yeandle6 in eccentric cell bodies of ommatidia in the compound lateral eye of Limulus. At low intensities of illumination, the fluctuations are maximal and occur infrequently. The higher the intensity of illumination, the greater the frequency of occurrence of the fluctuations and, in the steady state, the smaller their amplitudes. Furthermore, the amplitudes of the fluctuations vary markedly with the state of light and dark adaptation of the ommatidium.7' 8 After some time in the dark, the fluctuations elicited by lowlevel illumination of an ommatidium are large and distinct, but following a long exposure of the ommatidium to strong light, the amplitudes of the fluctuations become so small that they are barely discernible (Fig. 1).