Comments for the authors:
I have reviewed the manuscript ELE-00684-2021 entitled “Early queen infection shapes developmental dynamics and induces long-term disease protection in incipient ant colonies” written by Casillas-Pérez et al. The author's present an excellent set of experiments investigating how an early queen infection shapes the developmental dynamics of incipient colonies thorough the first year, by distinguishing the first solitary phase, after worker emergence (e.g., before and after hibernation) periods, as well as disease protection of workers in one year old ant colonies. I found the study to be quite interesting and the paper to be well-written. The authors have strong support for most of their conclusions, however, I do have a couple of concerns that would need to be addressed in a revision. For example, some details about the methodology are not entirely clear or missing, the experiments performed in the two different years are sometimes hard to distinguish, whereas the description of the statistical analyses in the SI can be simplified/shortened and better structured. Please see some general and specific comments below.

General comments
Introduction
1. Lines 117-126: Basically I think that the Introduction is well written, however, I suggest writing the aims in a simpler, straightforward manner (e.g. shorter sentences corresponding with the hypotheses of the study), and a summary of the main results can be also included at the end.

Materials and methods

1. Lines 130-131: suggest summing up separately what kind of experiments were performed in 2014 and 2018
2. Lines 136: it is not entirely clear to me the timing of the collection of the newly mated queens and their experimental exposure to a low dose of M. brunneum infective stages (conidiospores). If I understood well, these happened within the same week? The timing was similar between both years?
3. Lines 140: it would be more clear if the control treatment would be mentioned in a separate sentence.
4. Line 141: is this time period realistic for the study region?
5. Line 148: perhaps I missed something, but it is not clear to me what was the actual sample size after hibernation separately for the sham and pathogen treatments?
6. Lines 152-154: suggest giving the exact number of videos used in the different observation periods both for the sham and pathogen treatments.
7. Line 166: instead of “randomly sampled” suggest mentioning the actual sample sizes for the different time points.
8. Lines 171-172: suggest giving the sample sizes in different treatments separately for sham and pathogen treatments. Without them, the interpretation of the values in Line 175 becomes quite difficult. Although, maybe I have missed something, but in Line 175 the values given do not correspond with those from the SI (similarly within SI, the data from Lines 168-171 with those from Line 175). What is the reason? Moreover, suggest giving the outgrowth separately for the two pathogen treatment groups (high and very high number of conidiospores), as shown in the SI.
   Results
9. Generally, I suggest distinguishing more the results obtained based on the data collected in 2014 and 2018.
10. Lines 224: the actual results should be also included here.
11. Lines 254-257: these statistical analyses are not included in the Statistical analyses part.
12. Line 282: suggest being consistent with the denomination of the different variables through the MS. These can be emphasized by including the denomination of the variables in parentheses after their first mentioning (e.g. pathogen load in the M&M; and fungal load in the Results etc.).
13. Line 329: “intermediate challenge” was not used until now.
    Discussion
14. It would be nice for the discussion to start by briefly summarizing findings across experiments (in different years) with their relative contribution to the goals/hypotheses of the study.
15. Line 367: I am wondering why the queens still choose to rear more pupae under such circumstances and what can be the relationship between initial queen fat reserves, and the number of pupae reared? Suppose initial weight can determine the queen’s decision to lay more pupae, and this might have advantages under field conditions, issue that I think should be discussed more in detail.
    Figures
16. Figure 3. The significance levels are signalled with stars (but for example also in Figure 4c), whereas in Figure 2 the actual values are shown.

Specific comments
1. Line 49: Instead of “we here show” > “In this study, we show…”
2. Line 77: “et al.” is in italic thorough the MS
3. Line 90: the statement “the total duration of these effects may span only several weeks” should be underpinned with references.
4. Line 105: maybe it might be more accurate to write “affected” instead of “compounded”
5. Lines 106-109: I suggest completing the statement about relevancy with the fact that colonies of social insects are long-lived…
6. Line 219 (and elsewhere): “x” > “×”
7. Lines 226-228: “revealing a general trade-off in which…” > “revealing a general trade-off between …., where…”
8. Line 245: one “over” is in excess, please remove.
Supplementary information
1. Line 9: if the authors have the actual numbers of the collected queens (as described in the following sentence), I suggest writing those instead of “hundreds”.
2. Line 14: it is quite confusing that the authors refer to the results obtained in 2018, and then refer to previous work, and again to their results. It would be more straightforward if they would refer to the results of Pull & Cremer 2017 after they present their actual results related to the used samples (2018).
3. Lines 16-18: suggest reconsidering this sentence, e.g. “showed positive for” > “showed positive results for”
4. Lines 9-21: for the sake of better understanding suggest differentiating more between the results obtained in the two different years. The information from the Lines 41-43 can be also shown earlier (e.g. here) by specifying the details regarding the differences between the two years.
5. Lines 39-40: it should be also mentioned how long was the time period during which the temperature was changed from 4◦C to 23◦C. Based on the description one might think it was changed from one day to another.
6. Line 56: suppose an observation period lasted 3 days and the randomization was performed between the observation periods? These details should be clarified.
7. Lines 57-68: suggest also showing the exact number of pathogen and sham-treated colonies used in the different observation periods (e.g. in Figure S2), as it is hard to decipher for the readers how many colonies were actually analysed.
8. Line 75: “classifier output a” > “classifier output was a”
9. Lines 80-86: this sentence is too long and hard to follow. Suggest splitting into two separate sentences.
10. Line 81: “assigned the same identity” > perhaps “with the same”?
11. Line 95: based on the results given in the aggregation, I would not consider this metric “largely independent from colony size”, even if the r-value is low… Moreover, the details of these analyses are not mentioned in the Statistical analyses section.
12. Lines 96-97: I do not see how Figure S3 underpins the validity of the statement “This allowed us to compare ant behaviour in colonies of the different sizes across timepoints”…
13. Lines 99, 100: space id missing > “1 mm”, “1.3 mm2”
14. Lines 146-149: “x” > “×”
15. Lines 213-214: was not this a more correct approach also in the larvae and pupae? or instead of the time variable to include the factor (before and after hibernation)?
16. Lines 230-231: I think that colony size should be also included in the models (or the number of pupae, as the two are collinear), or a similar approach can be followed as in the models described in the followings (e.g. pupae-to-worker ratio). The argument against this (based on the repeated measures correlations), at least in the aggregation, does not hold, and the readers do not have sufficient details about how these correlations were performed. Moreover, I think that including before and after hibernation period instead of time (to avoid three way interactions or running separate models as performed in other cases) is also advisable.
17. Line 252: delete extra point after “queen”
18. Line 286: Instead of the function “coxph” I suggest using the function “coxme”, as it allows including colony as a random term.
19. Line 277-279: suggest transferring the information from the parenthesis into the main text, as it is highly relevant.

SI Tables
20. Suggest including an extra space between the Table rows, as they slur together. Moreover, the “treatment of queens” should be written out entirely (or to choose another abbreviation) instead of “(queenTx)” that hardens the interpretation of the interactions.
21. All in all, a better differentiation of the Table contents is advisable, whereas additional, explanatory information regarding the models should be included only in the text. Explanations regarding the predictors (if abbreviated) can be included instead.
22. Punctuation signs are missing, e.g. in Table S2: “We tested the three-way interaction between time treatment of queens (queenTx) and period before/after winter (W) against a model comprising each predictor and all two-way interactions.”

Comments for the authors:
I am very content with the revised submission and congratulate the authors on improving the manuscript by taking into account the initial feedback of the Reviewers and clarifying some important details. Most of my questions/concerns have been answered and I have only a handful of comments and suggestions left that should be addressed. The paper is a good contribution to its field of study and of interest to a wide audience and I think warrants publication in Ecology Letters.
General comments
Abstract
1. Line 51-52: suggest toning down or rephrasing the sentence part (“..but showed compensatory growth afterwards..”), as the trend was not significant, although the colonies were not different in size after one year. For example: “but the colonies were of similar size between treatments after one year of development”
2. Line 57: for those who read the paper it can be evident but for other readers, it can lead to misunderstandings, so I suggest emphasizing that the workers were challenged: “when challenged” > “when these were challenged”

Introduction
3. Lines 89: suggest deleting “single queen” as co-founded colonies can also occur…
4. Lines 96: “development superorganisms” > “development of superorganisms”
5. Line 104: “queens naturally encounter” > “queens naturally can encounter”
6. Line 106: “…we then tracked…” > “…we tracked…”
7. Line 111: “to study effects” > “to study the effects”

Materials and methods

1. Line 128: suggest removing “acute”, is a bit misleading as it is somehow contradicting “low levels of…”
2. Line 150: a comma is missing after “(Naiser et al. 2018)”
3. Lines 169-170: suggest reconsidering a bit this sentence, it is not entirely clear.
4. Line 182: “report” > “reported”
5. Line 198: delete extra comma
   Results
6. Line 216: there is an extra squared bracket > “…𝛸2=7.04, df=1, P=0.016])”
7. Lines 247-252: what were the effect sizes of these models? Suggest giving the effect sizes for all main models.
8. Lines 275-278: suggest reconsidering this fragment. It is hard to follow
9. Line 284: delete comma after “[LM]”, in other cases its not present
10. Line 307: the written “percent” can be deleted
    Discussion
11. Lines 343-344: similarly to comment 1…I find it more appropriate if written “colonies growing fast before winter slowed down after hibernation” (Lines 345-346)
12. Line 376: “this supports the notion a fecundity-immunity…” > “this supports the notion a fecundity-immunity…”
13. Line 407: delete extra comma
14. line 425: perhaps that they are less sessile can be also mentioned here
    Reference
15. The reference section still contains a few typos.

Figures
23. Figure 4c: R2 value is missing. Suggest presenting them (in this and also in other cases) also in the text.
24. Figures S1 and S5. In other figures, the exact p values are shown, on these only NS. These should be in line with the others.
Supplementary tables
25. Some of the supplementary tables need some adjustment > e.g. sometimes the content of a cell is slipped compared to others, space is missing, etc.