elementary charges into the cell is blocked (7). This amplification is explained by the cascaded reactions that link rhodopsin and the cGMP-activated channels in the surface membrane. Thus, the intense cGMP sink created by activation of PDE is sufficient to close a few hundred of the 7000 or so channels that are open at any instant in darkness. Further amplification results from the sizeable rate of ion flow through the channels themselves. The single photon response of cones is typically 10–100 times smaller than that of a rod and also considerably briefer. Given these functional differences it is perhaps not surprising that many of the transduction proteins are encoded by different genes in rods and cones (see ref.1).
Because it involves enzymatic mechanisms, visual transduction proceeds relatively slowly. In a monkey rod, for instance, the single photon response resembles the impulse response of a multistage low-pass filter with an integration time of about 0.2 s (7). This interval is comparable to the integration time of rod vision measured psychophysically (8), so that transduction itself, rather than subsequent processing in the eye or brain, apparently causes the poor temporal resolution of human rod vision. Although the single photon response of cones is too small to resolve, its average form can be inferred from the shape of the response to a dim flash. In primate cones it resembles the impulse response of a bandpass filter, with a delayed s-shaped rise to a peak and a prominent undershoot on the falling phase ( 9). The amplitude spectrum of the cone flash response has a peak at a frequency of 5–10 Hz, and the form of the amplitude spectrum resembles the psychophysically determined flicker sensitivity of human cone vision measured under light-adapted conditions (10).
Dark noise sets the ultimate limit on the performance of many devices that count photons, and retinal rods are no exception. The electrical noise of rods contains two dominant components that may be confused with photon responses: (a) occasional events resembling responses to single photons (the “discrete” component) and (b) a sustained fluctuation of smaller amplitude (the “continuous” component) (11). In a monkey rod the discrete noise events occur about once every 2.5 min (7). Psychophysical experiments indicate a similar magnitude for the rod “dark light,” the apparent rate of photon-like spontaneous excitations in dark-adapted rods (12,13). The temperature dependence of the rate of occurrence of discrete events gives the apparent activation energy of the process producing them as about 22 kcal mol−1 (1 kcal = 4.18 kJ) (11). This is close to the activation energy for thermal isomerization of 11-cis-retinal (14), suggesting that discrete events arise from thermal isomerization of rhodopsin's 11-cis-retinal chromophore. Additional evidence for the functional importance of thermal events is provided by behavioral experiments and recordings from retinal ganglion cells which show that the threshold for scotopic vision in toads is limited by a noise source with a very similar rate per rhodopsin molecule and temperature dependence (15). Although thermal activation occurs, it is infrequent: one calculates a 420-year average wait for a rhodopsin molecule at 37°C (7).
Cones are electrically noisier than rods, consistent with psychophysical evidence for a larger dark light in cones. In a monkey cone one component of the dark noise has a power spectrum like that of the cell's dim flash response (9). The photoisomerization rate that would produce an equivalent amount of noise is estimated as roughly 103 s−1. Bleaching a cone's pigment lowers the photon-like dark noise, suggesting that the noise may arise from thermal isomerization of pigment (9). Perhaps the red-shift in the absorption spectra of the pigments in red- and green-sensitive cones is inevitably accompanied by greater thermal instability (16).
The continuous noise of rods arises within the outer segment at a site in the transduction cascade downstream from rhodopsin (11), but the molecular mechanism has yet to be identified. The noise seems to result from shot effects of very small amplitude occurring at high frequency. The power spectrum of the continuous noise suggests that the shot effect is shaped by two of the four low-pass filter stages in an empirical quantitative model of the shape of the single photon response (11). Although the continuous component contrib-