Farm-Scale Effectiveness of Feed Additives Supplied through a Mineral Mix for Beef Cattle Grazing Tropical Pastures

Round 1

Reviewer 1 Report

Authors used:

a. different cattle breeds of different ages and size,

b. had different measuring of intakes (Site 1 : 75- and 90-day vs Site 2: 59 and 98 days) and 

c. sampling days (Site1 5 & 70 vs only one day on Site 2 : 60 day?) 

Line 19: Author talks of using 316 Nellore cattle but in lines 106 -108 says differently.

Line 71. The objective is not correct because talks of only Nellore cattle but used different breeds in the experiments reported. There is need to reformulate the objectives to cover all the scope in the manuscript.

Line 138: Are the eight animals per treatment or per site?

Line 144. How did you measure the proportion of animals considered as non eaters?  expand?

Line 148: Are the two ruminal samples per treatment or site?

line 151: Sampling of three times at site 1 vs one sample from site 2 why the imbalance?

Expt 2 used a mixture of bull breeds.

Author Response

Comments from Reviewer 1:

1. different cattle breeds of different ages and size,

*** Thank you. That is partially correct. We have used commercial Bos indicus “zebu” cattle which in Brazil are strongly represented by Nellore and Nellore crosses (i.e. more than 70%). We agree that ideally we would have used the same animal category across all experiments, with similar age and size, but this is many times not feasible. Despite the limitation, our main goal was achieved. To address the reviewer’s concern, we have made changes in ***Lines 19, 25, 106 and 107 to clearly indicate the common feature (zebu and zebu crosses).

1. had different measuring of intakes (Site 1 : 75- and 90-day vs Site 2: 59 and 98 days) and

*** Thank you for your concern. Despite monitoring intake on different experimental days, the same procedure was adopted using lithium sulphate in both sites. The reference for our method was given: [26] Dixon, R.M., Smith, D.R., Porch, I., Petherick, J.C. Effects of experience on voluntary intake of supplements by cattle. Australian Journal of Experimental Agriculture 2001 41, 581–592. https://0-doi-org.brum.beds.ac.uk/10.1071/EA00172

1. sampling days (Site1 5 & 70 vs only one day on Site 2 : 60 day?)

*** Again, thank you for your concern. We have noted the sampling days so readers can follow the procedures adopted. The actual dates are sometimes dictated by factors beyond our control (e.g. staff and/or students availability, weather events, etc). Again, the dates were indicated as they happened but are not necessarily the protocol to be followed. All scientific methods were properly cited and therefore can be found by readers interested in the subject.

Line 19: Author talks of using 316 Nellore cattle but in lines 106 -108 says differently.

*** This has been addressed. We’ve used 316 zebu and zebu-cross commercial cattle. In Brazil, this is strongly represented by Nellore and its crosses. See changes made in ***Lines 19, 25, 71, 106 and 107 to indicate that.

Line 71. The objective is not correct because talks of only Nellore cattle but used different breeds in the experiments reported. There is need to reformulate the objectives to cover all the scope in the manuscript.

*** Changes were made in ***Line 71 that now reads: “zebu and zebu-cross cattle grazing tropical pastures.”

Line 138: Are the eight animals per treatment or per site?

*** Thank you. We have added “per treatment” in ***Line 138.

Line 144. How did you measure the proportion of animals considered as non eaters?  expand?

*** The percentage of non-eaters was calculated based on the number of animals that had blood concentrations of lithium lower than 50 mg/L. This was based on the work of Kahn and properly referenced in ***Line 144: [27] Kahn, L.P. The use of lithium chloride for estimating supplement intake in grazing sheep: estimates of heritability and repeatability. Australian Journal of Agricultural Research 1994 45, 1731–1739. https://0-doi-org.brum.beds.ac.uk/10.1071/AR9941731. Please note that the analysis of the effect of treatment on the percentage of non-eaters was conducted using a generalized linear model for quasi-binomial distribution and Chi-square test. This was also indicated in ***Lines 237-238.

Line 148: Are the two ruminal samples per treatment or site?

*** Thank you. Two per treatment. ***Lines 148 and 149 now read: “from two random animals of each treatment”.

line 151: Sampling of three times at site 1 vs one sample from site 2 why the imbalance?

*** We've recorded numbers and days of sampling, acknowledging that actual dates may be influenced by factors beyond our control. While dates were reported as they occurred, they do not necessarily represent the prescribed protocol. The main thing is to note that scientific methods were appropriately cited for readers interested in the topic.

Expt 2 used a mixture of bull breeds.

*** Yes, zebu-cross cattle were used in Exp. 2. See changes made in ***Lines 19, 25, 71, 106 and 107 to address the differences. Thank you for your concern.  

We sincerely thank Reviewer 1 for dedicating time and effort to this manuscript. The revised version reflects significant changes based on your valuable input, evident throughout the entire document.

Reviewer 2 Report

The MS entiled “Farm scale effectiveness…” investigated the effects of salinomycin and virginiamycin delivered through MM on growing bulls grazing tropical pastures in two sites during different seasons and confirmed the effect on ruminal fermentation in vitro using sheep as rumen content donors. Results showed that virginiamycin in MM increased grazing beef bulls' ADG, with no clear link to rumen fermentation or coccidiostat effects.

General comments

Due to non-eaters in all treatments, can the authors make sure all the other animals’ intake is sufficient? How did the authors select the animals for the further detection? Does the Average daily gain were affected by the non-eaters? Why the bull does not consume the MM supplement, or what’s the possible reason? Will it affect the application of virginiamycin in the bull production? How to improve this situation? What’s the value of this study?

Specific comments

Pls list the parameters of the site1 and 2 such as the experimental date, animal number etc. in a table.

Use subtitles for each experiment specifically for experiment 2 and 3 such as the materials and methods, indexes be exanimated, and for the results.

Title: trough or through is right?

L21 ppm, change this unit.

L22: Over 123 days, cannot use the average day of the Site1 and 2

What’s the function of salinomycin and virginiamycin? pls have an introduction in the first part of the text.

L138 Eight animals in each unit?

Author Response

Comments from Reviewer 2:

The MS entiled “Farm scale effectiveness…” investigated the effects of salinomycin and virginiamycin delivered through MM on growing bulls grazing tropical pastures in two sites during different seasons and confirmed the effect on ruminal fermentation in vitro using sheep as rumen content donors. Results showed that virginiamycin in MM increased grazing beef bulls' ADG, with no clear link to rumen fermentation or coccidiostat effects.

General comments

Due to non-eaters in all treatments, can the authors make sure all the other animals’ intake is sufficient? How did the authors select the animals for the further detection?

***Thank you. The percentage of non-eaters was determined by inference from animals with blood lithium concentrations below 50 μg/L, following the method established by Kahn and appropriately referenced. Like most, if not all, methods, there are limitations. It works as a spot sample; therefore, we cannot state the intake of all other animals was sufficient, however, we can make conclusions from evidence and reasoning rather than from explicit statements. Animals were selected randomly from each treatment.

Does the Average daily gain were affected by the non-eaters? Why the bull does not consume the MM supplement, or what’s the possible reason? Will it affect the application of virginiamycin in the bull production? How to improve this situation? What’s the value of this study?.  

***Yes, the average daily gain would be affected but that’s why we used a number of replicates large enough to overcome this issue. An animal can refuse to eat a supplement for a number of reasons, e.g. behaviour, palatability, bullying, The main concern addressed in this work is that monensin, one of the most studied feed additives, is known to reduce intake. When this ingredient is added to a mineral mix of low consumption, it could drastically affect the supplement intake, hence why virginiamycin was introduced in the market. This study was designed to validate if the use of different additives could “improve this situation” as you point out. We believe that this work has great value since it is testing commercial products, ready for adoption.

Specific comments

Pls list the parameters of the site1 and 2 such as the experimental date, animal number etc. in a table.

***Thank you for the suggestion. We opted to keep these in the text and leave Tables for results and/or distinctive attributes of experiments and/or sites.

Use subtitles for each experiment specifically for experiment 2 and 3 such as the materials and methods, indexes be exanimated, and for the results.

***Do you mean we should use a shorter word “exp.” instead of the whole word “experiment”? We have now done this throughout the manuscript where found appropriate.

Title: trough or through is right?

***Thank you very much. You are absolutely right. We meant “through” a mineral mix.

L21 ppm, change this unit.

***Thank you very much. We see no reason for debate over the units ppm or mg/kg, as they are synonymous and convey the same measurement. The reason we chose mg/kg is because it’s deemed the most accepted and used in the countries where the authors reside and work.

L22: Over 123 days, cannot use the average day of the Site1 and 2

***The message we tried to convey was that the experimental animals were grazing tropical C4 grasses, that have distinct characteristics to differentiate them from a temperate C3 grass, such as rye grass for example. To clarify this, the following change has been made in ***Lines 22 and 23 that now read: “these bulls grazed tropical grasses on pastures of guineagrass, palisadegrass, or bermudagrass.”

What’s the function of salinomycin and virginiamycin? pls have an introduction in the first part of the text.

***The following has been added to ***Lines 54-56: “Ionophores have a tendency to impact the consumption of MM, potentially limiting mineral intake and making the intended ionophore dosage impractical” and as stated on ***Lines 71-73: “the objective of this study was to evaluate the growth-promoting effects of both the ionophore salinomycin and the non-ionophore virginiamycin delivered through MM supplementation”.

L138 Eight animals in each unit?.

***Thank you. We have added “per treatment” in ***Line 138.

We appreciate the thoughtful considerations and suggestions from Reviewer 2 that have undoubtedly enhanced the overall quality of the manuscript.

Reviewer 3 Report

Dear Authors,

Overall, this manuscript lacks clarification on data collection and statistical analysis, furthermore, the writing could be improved to improve the reading flow. Many sentences lack scientific formal writing for a peer-reviewed manuscript, such as Line 46: “challenge of administering them effectively.” I believe this sentence could be rewritten for better clarification. What would be effective? Assure continuous consumption? Daily intake? Mixture with the mineral mixture?

There are definitely some improvements that could be made to the writing of this manuscript. However, the experimental design for all experiments within this manuscript does not show appropriate application. Experiment 1: Missing fixed effect of location. Additionally, the way that samples were collected does not reflect the true experimental unit for this experiment. For example, lithium was used to estimate MM intake; were the 8 animals removed from each paddock or each block? How many animals were in each paddock? There is no clarification on that matter. Figure 1 shows an n of 240, however, there were 16 paddocks in location 1 and 8 paddocks in location 2. If the intake was measured in 8 animals per paddock, shouldn't it be n =192? The sample size in each table (i.e., s1) is confusing to understand and needs some clarification. 

Experiment 2: More information could be provided in the material and methods. It is not clear what was the design of this experiment.

Experiment 3 has no negative control. Were there any differences in the in vitro data due to salinomycin addition at all? There is no way to prove that. What were the blocks utilized in this CRBD design? A lot of information is missing. 

Vocabulary is scarce and repetitive throughout the text. There is a clear space for improvements.

Author Response

Comments from Reviewer 3:

Dear Authors,

Overall, this manuscript lacks clarification on data collection and statistical analysis, furthermore, the writing could be improved to improve the reading flow. Many sentences lack scientific formal writing for a peer-reviewed manuscript, such as Line 46: “challenge of administering them effectively.” I believe this sentence could be rewritten for better clarification. What would be effective? Assure continuous consumption? Daily intake? Mixture with the mineral mixture?

***Thank you for your concern. We have attempted to be as clear as possible regarding both data collection and statistical analysis. We have documented throughout the manuscript the actual collection and sampling numbers and days, recognizing that actual dates may be influenced by external factors. It is important to note that scientific methods were appropriately cited for readers interested in the topic. Despite this, we are in agreement that improvements can be made to improve the quality of the final manuscript and the reading flow. Please see some modifications indicated below. Other suggestions from different reviewers can be found in the revised manuscript in this new submission.

***Lines 46-47 now read: “the challenge of administering them effectively to ensure sufficient mineral consumption.”

***In addition, we have re-written the description of the statistical analysis to improve clarification as suggested. The full paragraph in ***Lines 235-253 now reads:

“All statistical analyses were conducted using the open-source software R [37] and data was assessed prior to analysis for normality and homoscedasticity and, if necessary, data was transformed according to the box-cox procedure [40]. For all analysis of Exp 1 and 2, the paddocks were considered experimental units and the average of the measurements of the bulls in each paddock was utilized for analysis [38]. Average daily gain, MM intake, ruminal SCFA, and oocysts counting of Exp 1 as well as ADG and MM intake of Exp 2 were analysed using the linear mixed models’ procedure of the package “nlme” [39]. The models included treatment as fixed factor and block within location as random. The percentage of non-eaters was calculated based on the number of animals, which presented lithium blood concentrations lower than 50 mg/L [27]. The analysis of the effect of treatment on the percentage of non-eaters was conducted using a generalized linear model for quasi-binomial distribution and Chi-square test. In Exp 3, the effect of time of exposure for the MM of salinomycin on rumen parameters was evaluated using a generalized linear model. Pearson correlation analysis was conducted on data from the individual intake of Exp 1 between the two collection dates (i.e., Intake 1 and 2). For the Pearson correlation analysis the experimental unit was considered to be each bull and not the paddock. Differences were declared significant at p ≤ 0.05, and trends were declared at p ≤ 0.10.”

There are definitely some improvements that could be made to the writing of this manuscript. However, the experimental design for all experiments within this manuscript does not show appropriate application. Experiment 1: Missing fixed effect of location. Additionally, the way that samples were collected does not reflect the true experimental unit for this experiment. For example, lithium was used to estimate MM intake; were the 8 animals removed from each paddock or each block? How many animals were in each paddock? There is no clarification on that matter. Figure 1 shows an n of 240, however, there were 16 paddocks in location 1 and 8 paddocks in location 2. If the intake was measured in 8 animals per paddock, shouldn't it be n =192? The sample size in each table (i.e., s1) is confusing to understand and needs some clarification.

***The effect of location was taken into account in our statistical models as a random factor with blocks nested within. The decision to modelled location as a random and not fixed factor was due to the fact that were infinite possibilities of location and in this case, we were not interested in this effect itself but in controlling for the random variation associated with it.

***To further clarify the allocation of animals, the following has been added to ***Lines 112-117: “Animals were allocated to treatments based on liveweight rankings at the start of each trial. The number of animals in each paddock varied according to the size of the paddock and the forage mass available. In average, there were 7 to 8 animals per paddock in Site 1 and 24 to 25 in Site 2. The stocking rate of each paddock was calculated in order to maintain the sward heights accordantly to the recommended grazing height for each pasture species.”

Experiment 2: More information could be provided in the material and methods. It is not clear what was the design of this experiment.

We have re-written the materials and methods section of Experiment 2 in ***Lines 173-178 that now read: “The Exp 2 was conducted at the same set of paddocks of site 2 in Exp 1 in a completely randomised block design with the two treatments distributed into four blocks where each block included one repetition (i.e. paddock) per treatment. The paddock was considered the experimental unit; the same set of paddocks of site 2 in Exp 1, but now the 8 paddocks were distributed into these 4 blocks. It lasted 133 days from January to May 2010, and treatments were:….”

Experiment 3 has no negative control. Were there any differences in the in vitro data due to salinomycin addition at all? There is no way to prove that. What were the blocks utilized in this CRBD design? A lot of information is missing.

***The negative control, with no addition of additive, was left out of this particular analysis because we’ve analysed the effects of mixture time. Since the additive was not mixed in, we left it out.

Comments on the Quality of English Language

Vocabulary is scarce and repetitive throughout the text. There is a clear space for improvements.

***Thank you for your concern. We genuinely appreciate Reviewer 3's time and dedication to this manuscript. The revised version incorporates substantial changes inspired by your valuable input, evident across the entire document. Hopefully the “space for improvements” has been filled as a reflection of the feedback from all Reviewers.

Reviewer 4 Report

Overall, interesting work that explores an important issue in beef cattle production. 

In the title of the paper, the authors may mean to use the word "through" instead of "trough". If "trough" is intended, then the title should be reworded to improve the reader's understanding of the title. This issue also occurs with the word "trough" used in line 372 of discussion when the word "through" should be used instead.

The results presented in the this manuscript describe bulls. These results may or may not be generalizable to cows on pasture. Behavior among bulls yearling bulls on pasture may be different than mature cows on pasture. 

In section 2.1.2, it is not clear how many animals were included in each paddock. It is also not clear in animals were randomly allocated to each group. What were the animals stratified by, and were they randomly allocated to groups such that each group had various ages, breeds, and weights, or similar ages, breeds, and weights.

Only 8 of the animals were selected for lithium sulphate sampling. The authors include no statistical justification for selection of 8 animals (line 138). Were any sample size calculations conducted that indicated 8 animals would be sufficient to detect a meaningful difference between groups? Similarly, no sample size justification was given for why the 2 animals per group (line 148) or the 5 animals per group (line 161) were selected for ruminal samples. I suggest the authors provide a justification for why these numbers of animals were selected (e.g., budget constraints, sample size calculations, etc.). 

How were the doses (mg/kg) of salinomycin and virginiamycin chosen for Experiments 1 and 2?

The hypothesis and objectives of Experiment 3 is not clear. Please describe the hypothesis and objectives more clearly. For example, I believe you intended to test the effectiveness of salinomycin after it had been mixed with the MM and left in the trough for different amounts of time. This was not clear when I read section 2.3 initially. 

In lines 318 to 324 of the discussion, you state that the increase in the concentration of soybean meal from 5% to 15% was responsible for the increase in intake? This increase in intake could also be due to a lower dose of salinomycin included in the SHI treatment vs. the SLI treatment. Please discuss if differences in palatability of salinomycin could explain differences in intake. This potential palatability issue could also explain why there were so many non-eaters in the study?

Lastly, does the difference in ADG between the CON and VGN treatment groups in Experiment 2 justify the cost of the VGN treatment to producers? It seems as though the difference in ADG was small, and although statisically significant, it may not be practical due to the increase in cost associated with the addition of virginiamycin to the diet?

Overall, the English is good, with a minor correction to sentence structure needed in the sentence that begins on line 314 and ends on line 316. I believe this is a run-on sentence, that should be divided into two sentences to improve clarity (period after "systems", then beginning the new sentence with "Validating MM as a carrier...").

Author Response

Comments from Reviewer 4:

Overall, interesting work that explores an important issue in beef cattle production.

In the title of the paper, the authors may mean to use the word "through" instead of "trough". If "trough" is intended, then the title should be reworded to improve the reader's understanding of the title. This issue also occurs with the word "trough" used in line 372 of discussion when the word "through" should be used instead.

***Thank you very much. You are absolutely right. We meant “through” a mineral mix. The correction was also made in ***Line 378 that now reads: “additives supplied through a MM for beef cattle grazing tropical pastures”.

The results presented in the this manuscript describe bulls. These results may or may not be generalizable to cows on pasture. Behavior among bulls yearling bulls on pasture may be different than mature cows on pasture.

***You are absolutely right. Behaviour amongst bulls is quite unique. Brazil is different to Australia and US for example, where males are usually castrated and some sort of HPG is normally used. The latter is illegal in Brazil, hence why the natural hormones of non-castrated males are the preferred option of management. The following paragraph has been added in ***Line 374-377: “Behavioural patterns in young bulls grazing on pasture may exhibit distinctions compared to other animal categories in the same environment and the current findings may or may not be applicable in the broader context. However, the relevance of the animal category studied to commercial operations is clear.”

In section 2.1.2, it is not clear how many animals were included in each paddock. It is also not clear in animals were randomly allocated to each group. What were the animals stratified by, and were they randomly allocated to groups such that each group had various ages, breeds, and weights, or similar ages, breeds, and weights.

***Thank you very much. The following has been added in ***Lines 112-117: “Animals were allocated to treatments based on liveweight rankings at the start of each trial. The number of animals in each paddock varied according to the size of the paddock and the forage mass available. In average, there were 7 to 8 animals per paddock in Site 1 and 24 to 25 in Site 2. The stocking rate of each paddock was calculated in order to maintain the sward heights accordantly to the recommended grazing height for each pasture species.”

Only 8 of the animals were selected for lithium sulphate sampling. The authors include no statistical justification for selection of 8 animals (line 138). Were any sample size calculations conducted that indicated 8 animals would be sufficient to detect a meaningful difference between groups? Similarly, no sample size justification was given for why the 2 animals per group (line 148) or the 5 animals per group (line 161) were selected for ruminal samples. I suggest the authors provide a justification for why these numbers of animals were selected (e.g., budget constraints, sample size calculations, etc.).

***Thank you. Yes, prior to the start of the experiments we had conducted a power analysis using the coefficient of variation expected for the technique(s) and differences expected amongst treatments to calculate the requirements for animal numbers (sample size calculations as you suggested). The animals were selected randomly, but we had not previously indicated the procedure. Now in ***Lines 146-147 of the revised copy of the manuscript read: “Eight animals were selected randomly per treatment to be removed…”

How were the doses (mg/kg) of salinomycin and virginiamycin chosen for Experiments 1 and 2?

***The doses utilised were correctly referenced in ***Lines 95-96: “The concentration of additives in the MM was defined expecting an intake of 30 mg/100 kg LW for salinomycin and virginiamycin [12,20,1].”

The hypothesis and objectives of Experiment 3 is not clear. Please describe the hypothesis and objectives more clearly. For example, I believe you intended to test the effectiveness of salinomycin after it had been mixed with the MM and left in the trough for different amounts of time. This was not clear when I read section 2.3 initially.

***Perfect. Thank you! Now in ***Lines 208-210 it reads: “The objectives were to assess the efficacy of salinomycin following its combination with MM, and subsequent exposure to varying durations within the trough.”

In lines 318 to 324 of the discussion, you state that the increase in the concentration of soybean meal from 5% to 15% was responsible for the increase in intake? This increase in intake could also be due to a lower dose of salinomycin included in the SHI treatment vs. the SLI treatment. Please discuss if differences in palatability of salinomycin could explain differences in intake. This potential palatability issue could also explain why there were so many non-eaters in the study?

***The higher inclusion of soybean meal affected the MM intake because the animals were more prone (attracted) to consume the supplement with higher inclusion. The main observation was that the intake of the active ingredient did not differ across treatments. ***Lines 338-342 read: “the strategy of increasing soybean meal concentration from 5 to 15% for the SHI group was effective to increase the MM intake; however, because the salinomycin concentration was reduced by 60% in that MM, the active ingredient intake was the same for both SLI and SHI treatments.”

Lastly, does the difference in ADG between the CON and VGN treatment groups in Experiment 2 justify the cost of the VGN treatment to producers? It seems as though the difference in ADG was small, and although statisically significant, it may not be practical due to the increase in cost associated with the addition of virginiamycin to the diet?

***That’s a very good point, which needs to be considered. We were going to leave this up to readers, but since you brought up, we decided to add the following paragraph in ***Lines 379-382: “For any additive, it’s important to consider that despite possible positive impact on cattle performance, it is essential to evaluate the cost of the strategy, emphasizing the need for a balanced evaluation that takes both effectiveness and economic factors into account.”

Comments on the Quality of English Language

Overall, the English is good, with a minor correction to sentence structure needed in the sentence that begins on line 314 and ends on line 316. I believe this is a run-on sentence, that should be divided into two sentences to improve clarity (period after "systems", then beginning the new sentence with "Validating MM as a carrier...").

***Thank you very much. Your thoughtful and constructive feedback has significantly enhanced the quality of the work. We appreciate the time and effort they dedicated to providing insightful comments, which have been instrumental in refining and improving the overall content.

Round 2

Reviewer 3 Report

Some improvements were made, but it still needs some clarification.

Line 19-21: "In Exp. 1, 316 zebu (Bos indicus) Nellore bulls [225 ± 26.7 kg live weight (LW)], were randomly allocated to four treatments: 1) MM no additives (CON; n = 6), 2) MM with salinomycin at 1,950 mg/kg (SLI; n = 6), 3) MM with salinomycin at 780 mg/kg (SHI; n = 6), and 4) MM with virginiamycin at 1,950 mg/kg (VGN; n = 6)."

The "n" standard for? There is no mention of paddocks in the abstract. If the experimental unit is the paddock, why is it not included here?

Line 243-145:"All statistical analyses were conducted using the open-source software R [37] and 243 data was assessed prior to analysis for normality and homoscedasticity and, if necessary, 244 data was transformed according to the box-cox procedure [40]." was any data transformed ? If yes, clarify which.

"The effect of location was taken into account in our statistical models as a random factor with blocks nested within. The decision to modelled location as a random and not fixed factor was due to the fact that were infinite possibilities of location and in this case, we were not interested in this effect itself but in controlling for the random variation associated with it."

Clarify the model in the statistical analysis section. What was the exact random statement in the model? Where does the paddock enter this model then?

Line 144-147: the lithium provision was done for only 10 hours prior to access blood lithium concentration and that was used to classify the "non-eaters"? This section needs more clarification and discussion. The discussion section is lacking a link of the results. In one section, it is mentioned the importance of mineral consumption, whereas, in the other section, it is indicated that animals supplemented with VM had a lower mineral intake, a greater percentage (numerically) of animals that did not consume the mineral but gained more weight. The utilization of this lithium procedure might only indicate the percentage of animals that did not consume the mineral supplement in the hour period. Please clarify it in the text. Further, elaborate on the possible mechanism by which the authors believe that VM supplementation might be improving ADG it is not clear in the current version."

“***The negative control, with no addition of additive, was left out of this particular analysis because we’ve analysed the effects of mixture time. Since the additive was not mixed in, we left it out.”

What is the null hypothesis of this analysis? If you are trying to show that salinomycin addition is not compromised over time you should include the control treatment in the analysis. If you are not able to show that the addition of salinomycin using this technique was not different from the treatment containing no salinomycin how do you expect to prove that salinomycin was effective after 300 days? You are omitting data that proves it was effective at all. Please make sure you have the correct treatment sets to prove your null hypothesis.

None.

Author Response

Comments from Reviewer 3:

Some improvements were made, but it still needs some clarification.

***We have addressed your concerns as indicated in the responses below.

Line 19-21: "In Exp. 1, 316 zebu (Bos indicus) Nellore bulls [225 ± 26.7 kg live weight (LW)], were randomly allocated to four treatments: 1) MM no additives (CON; n = 6), 2) MM with salinomycin at 1,950 mg/kg (SLI; n = 6), 3) MM with salinomycin at 780 mg/kg (SHI; n = 6), and 4) MM with virginiamycin at 1,950 mg/kg (VGN; n = 6)."

The "n" standard for? There is no mention of paddocks in the abstract. If the experimental unit is the paddock, why is it not included here?

*** The n conventionally refers to the sample size expressed as the number of replicates regardless of whether the experimental unit is an animal or a paddock. In our case, the paddock was chosen to provide more robustness to the results based on recommendations for grazing experiments by Fisher, (2000). We had originally included the n number in the abstract to provide more information to readers, but the word limits suggested by the journal is 200 (and our had about 260 after adding other considerations from reviewers, therefore we have decided to remove the “n’s” from the abstract. Despite this, the reader will still be able to find this detailed info on MM on ***Lines 21-23 and 28-29.

Line 243-145:"All statistical analyses were conducted using the open-source software R [37] and 243 data was assessed prior to analysis for normality and homoscedasticity and, if necessary, 244 data was transformed according to the box-cox procedure [40]." was any data transformed? If yes, clarify which.

"The effect of location was taken into account in our statistical models as a random factor with blocks nested within. The decision to modelled location as a random and not fixed factor was due to the fact that were infinite possibilities of location and in this case, we were not interested in this effect itself but in controlling for the random variation associated with it."

Clarify the model in the statistical analysis section. What was the exact random statement in the model? Where does the paddock enter this model then?

Thak you

***Thank you again for your concern. We have rearranged the statistical analysis section aiming at making it clearer. Please note that we have included the description of all variables that were transformed and the type of transformation applied in the statistical analysis section. The latter can be found in ***Lines 389-392. As for further clarifications to our model applied for exp1, we have described it in detail in ***Lines 392-393. Again, note that paddock was the experimental unit in that experiment and therefore not included as a factor in the statistical model. 

Line 144-147: the lithium provision was done for only 10 hours prior to access blood lithium concentration and that was used to classify the "non-eaters"? This section needs more clarification and discussion. The discussion section is lacking a link of the results. In one section, it is mentioned the importance of mineral consumption, whereas, in the other section, it is indicated that animals supplemented with VM had a lower mineral intake, a greater percentage (numerically) of animals that did not consume the mineral but gained more weight. The utilization of this lithium procedure might only indicate the percentage of animals that did not consume the mineral supplement in the hour period. Please clarify it in the text. Further, elaborate on the possible mechanism by which the authors believe that VM supplementation might be improving ADG it is not clear in the current version."

***The reviewer is correct. We stated this 10-h period (from 6h to 16h) in ***Line 150 from the MM section. We followed the methodology according to Dixon et al. (2001). However, we agree that it would be worth exploring this more in the discussion section. In this revised submission, this can be found in ***Lines 345-350, 377-378, 397-399 to allow greater depth in the discussion of this topic.

Regarding the intake and ADG of animals receiving VGN, we stated that “Animals from VGN did not decrease their MM intake compared to CON group in exp 1, but there was a trend in exp 2. There is no evidence in the literature reporting decreased MM intake due virginiamycin inclusion [20,42]. This trend observed in exp 2 may be explained by the higher concentration of active ingredient in the MM containing virginiamycin on exp 2 compared to exp 1 (2,522 vs. 1,950 mg/kg)”. The latter is found in ***Lines 374-378, and mot necessarily reflects a lower mineral intake, but rather a greater percentage (numerically) of animals that did not consume the mineral but gained more weight. To clarify, the only lower intake (p ≤ 0.05) was for SLI in Exp 1.

“***The negative control, with no addition of additive, was left out of this particular analysis because we’ve analysed the effects of mixture time. Since the additive was not mixed in, we left it out.”

What is the null hypothesis of this analysis? If you are trying to show that salinomycin addition is not compromised over time you should include the control treatment in the analysis. If you are not able to show that the addition of salinomycin using this technique was not different from the treatment containing no salinomycin how do you expect to prove that salinomycin was effective after 300 days? You are omitting data that proves it was effective at all. Please make sure you have the correct treatment sets to prove your null hypothesis.

The objective of exp 3 was to assess if the exposure time of salinomycin to the MM mix would affect the in vitro gas production. The reason we haven’t included the control treatment without salinomycin initially is because the null hypothesis in this trial was that exposure time would not affect in vitro gas production. Our understanding was that, since we were not testing the effect of the inclusion of salinomycin itself, the inclusion of the “second control” (i.e. without salinomycin) would not be necessary. However, given the recommendations of the reviewer we have reanalysed the results including the control group without salinomycin and re-written the results (***Lines 619 - 623) and discussion sessions (***Lines 783 - 787).