Referee 2'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.
**********

Comments on Guyader and Burch
In this manuscript the authors have nicely incorporate the diet composition model in optimal foraging theory to explain the observed specialist and generalist bacteriophages found in marine environment.

I only have a few comments on this manuscript.

(1). p.3, the last fourth line, the second "e.g." should be omitted

(2). p. 9, it is not clear on how the lysis time is determined. For example, what are the time intervals between each sampling? Or maybe there are some preliminary experiments done to determine the best sampling intensity around the suspected lysis time? It would seem to have introduce a large error if the sampling interval is large.

(3). p.9, Is there a reason why the burst size was determined at time t = 2L? Is this just an arbitrary decision? I think it may be clearer for the authors to include an experimentally determined one-step growth curve, preferably one like Fig. 1, for the readers to visualize the infection process.

(4). p. 9, the bottom third. The authors stated that 45 min at 37 C would allow the phages to have two full cycles (by the way, the "2" may need to be changed to "two"). Has this been experimentally demonstrated? Again, an experimentally determined one-step growth curve can tell us.

(5). p.9, the bottom third. The authors stated that adding chloroform would destroy bacterial cells and release all fully formed, matured phage particles. I think the authors are right on the first half of the sentence. Addition to bacterial culture infected by small phages (ssRNA and ssDNA phages) will not release any trapped phage progeny inside the cell. The chloroform can only work for large phages like lambda, T4, T7, etc. The simple explanation is that the small phages do not accumulate phage-encoded endolysin, therefore, destruction of cell membrane (by chloroform), while kills the cell, does not release the endolysin, thus, no cell lysis.

(6). p. 9, the bottom third. The authors used a ratio of 3.35:1 of E. coli and S. typhimurium for phage plating. I assume this skewed ratio is because of different growth rate of these two hosts. Is that the reason? It would be nice for the authors to explain it so other readers may appreciate the nuance of it.

(7). p. 12, middle, about Figure 2. How was Figure 2 generated? Can the authors show an example on how the calculations were made? I've tried to calculate it using the given parameter values to no avail. Also, on Fig. 2, the y-axis used is ln(growth rate). At host density of 10^8, the y-axis corresponds to around 13. Do the authors mean that the growth of phiX174 is exp(13) = 442413.392? That seems to be unreasonably high.

(8). p. 14, last paragraph. There should be an indent at the beginning of the paragraph.

(9). p. 15, about Fig. 5A and B. Again, I've tried to generate these Figures to no avail.
An example on how this was done would be great.

(10). p. 15, Discussion, last line. The authors stated that the G4 was advantageous at a high host density and phiX174 was advantageous at a low host density. Where does this statement come from? I flipped through the Results section and can't seem to find the relevant section that leads the authors to come to this conclusion. Is it the Fig. 3 the authors referring to? I can see why at low host density it is good to be phiX174, but I don't see the result demonstrating that it is good to be G4 when the host density is high.