Referee 1's Review:

**********
N.B. These are the comments made by the referee when reviewing an earlier version of this paper. Prior to publication, the manuscript has been revised in light of these comments and to address other editorial requirements.
*********

General

This is a carefully performed comprehensive study of visual transduction in locust photoreceptors. The paper addresses an important question of whether the coding of visual information is adapted to statistical properties of natural stimuli. The authors compare responses to white noise stimulation with those to naturalistic stimuli with 1/f power spectrum at different temperatures and intensities of background light. The main conclusion of the paper is that there might be a preference for inputs with 1/f statistics. The paper is well written and I have only few suggestions for improvement.

Major comment.

The authors suggest that small non-linearity in photoreceptor responses allow improvement of coding of naturalistic stimuli. I would like to see more explanations to this suggestion. 1) Where non-linear system is considered, the encoding can be better understood in the time then in the frequency domain. Can you explain in the time domain encoding of which stimuli is improved? 2) Alternatively, the non-linearity can be a consequences of constraints imposed on coding of light stimuli and adaptation to light. Can you also discuss this alternative view and present some arguments pro and contra the hypothesis of adaptive importance of the non-linearity.

Minor comments

P. 8. "Because we believe that intrinsic functional variability is an important feature of locust vision, we do not show averaged quantities" Can you give descriptive statistics, which would also reflect the variability?
P. 10 "For naturalistic stimulation .... the data are presented using arbitrary units". I think the data should be presented using the same units as the white noise stimulation. Otherwise these two conditions cannot be compared. Please explain how the contrast used in white noise stimulation can be related and compared to NS stimulation.
P. 11. "The signals were sampled at 1 (WN) or 10 (NS) kHZ" Why did you use different sampling fro two conditions?
P. 18. "...resulting in a distribution close to the one seen at the dark adapted state". Distribution of what? Please be more precise. Do you mean the variation of response?
P. 19. " ....for nine relevant parameters as defined in Materials and Methods" The reader is unlikely to remember what are these parameters.
P. 21. Bump noise analysis. Since bumps are described by gamma-distribution, their shape can vary. Does the shape of bumps vary? Can you present the data of variability of all parameters of bumps?
P. 26. Non-linear component enhance 'interesting' patterns in the stimulus. Can you give more details on such enhancing. Please, explain why such patterns are 'interesting'
P. 26. "Non-linear amplification seems to be really specific of certain patterns". Can you be more specific?
P. 28. "The elementary events do not depend on membrane time constant, but depend on the number of open channels (i.e. resistance). Since the time constant is proportional to the resistance, this phrase does not make sense.
P. 31. "Thus, it seems that the membrane helps to determine the speed at which ... but does not limit the speed of transduction reactions". I do not understand this statement.

Discussion
P.31. "This encoding task is helped by the reflectance .... and by relative movements". This phrase does not make sense for me.