Organosolv and Hydrothermal Pretreatments of Sugarcane Bagasse and Straw and Enzymatic Hydrolysis of Hemicellulosic Liquor
Round 1
Reviewer 1 Report
Comments and Suggestions for Authors
The authors have already followed the NREL procedure for compositional analysis of raw and pretreated biomass, so why are they finding holocellulose content, alpha cellulose, hemicellulose, AIL, and soluble lignin using another procedure?
The DNS method is not an accurate assay for determining the sugars in pretreatment liquor, as it contains furans (furfural and 5-HMF), which also react with the DNS reagent to give overestimated results compared to the actual. Authors may find such studies in the literature.
HPLC is the best way to analyze the compounds in hydrolysates.
Authors should check the trade name of the enzyme cocktail, Cellic CTec3 HS, which has cellulase and hemicellulase activities. It's a blend.
It would be good to provide the mass of the biomass after hydrothermal and organic solvent extraction. Also, representing the results in terms of g in Table 1 would be suitable for better understanding, instead of %.
The sentence in line 273 is not clear.
Line 283: Mass loss of 8.79% in SBH? In Table 1, hydrothermal pretreatment greatly removed hemicellulose. Please check this data.
Figures 2 and 3 are redundant, as the composition data before and after pretreatment are already shown in Table 1. Reviewers and general readers can easily see the differences in cellulose, hemicellulose, and lignin before and after the pretreatment.
Lines 302 to 304: To understand the delignification of SB and SS from the selected pretreatments, it is essential to show the mass recovery after the pretreatment. For example, after hydrothermal pretreatment, 8.5 g of biomass was recovered from 10 g of pretreatment. In 8.5 g, what percentage of cellulose, hemicellulose, and lignin should be provided in the table for ease of understanding?
Lines 389 to 394: Several studies on hydrothermal pretreated biomass use cellulase and hemicellulase together or hemicellulose alone, achieving more than 80% hemicellulose hydrolysis. For the authors' information, after the hydrothermal pretreatment, enzymatic hydrolysis was conducted without separating the liquor, reaching more than 80% hemicellulose hydrolysis. Why did the authors state an adverse effect on hemicellulose due to the presence of lignin, etc.? Please check the literature. Has the author used the wrong enzyme for hemicellulose hydrolysis?
There are several articles published on the same topic that are more advanced than this work, including hydrothermal pretreatment without separating the liquors, using cellulase and hemicellulase for hydrolysis of cellulose and hemicellulose with a lower concentration of furfural and 5-HMF.
The authors provided a significant elevation of the study in the introduction, but nothing new in the materials and methods, results, and discussion.
Author Response
Comments 1. The authors have already followed the NREL procedure for compositional analysis of raw and pretreated biomass, so why are they finding holocellulose content, alpha cellulose, hemicellulose, AIL, and soluble lignin using another procedure?
Response 1. Thank you for your comment. I would also like to take this opportunity to clarify that the NREL protocol was used for the compositional analysis of raw and pretreated biomass, providing global quantification of cellulose, hemicellulose, and lignin. Furthermore, we included methods for determining holocellulose and α-cellulose, as well as quantifying insoluble and soluble lignin, to obtain a more detailed characterization and validate the results obtained.
These classic methods, such as the Klason method and holocellulose/α-cellulose analyses, allowed for a more precise discrimination of the different carbohydrate and lignin fractions. Furthermore, they were useful for discussing structural aspects, such as cellulose purity and crystallinity, as well as for establishing comparisons with traditional studies in the field of plant fibers. Thus, the combination of procedures constitutes a complementary strategy that expanded the level of detail in the characterization and provided greater robustness to the data.
Comments 2. The DNS method is not an accurate assay for determining the sugars in pretreatment liquor, as it contains furans (furfural and 5-HMF), which also react with the DNS reagent to give overestimated results compared to the actual. Authors may find such studies in the literature.
Response 2. Regarding the DNS method, we recognize that, although it is widely used in lignocellulosic biomass studies, the results obtained from this analysis end up overestimating the actual sugar concentration due to the interference of compounds such as furfural and 5-HMF present in the pretreatment liquors. Therefore, we used DNS only as an auxiliary method and emphasized HPLC quantification, which provided more specific and reliable results. The text was revised to clarify this distinction.
Commnets 3. HPLC is the best way to analyze the compounds in hydrolysates.
Response 3. See the answer to the previous question.
Comments 4. Authors should check the trade name of the enzyme cocktail, Cellic CTec3 HS, which has cellulase and hemicellulase activities. It's a blend.
Response 4. We corrected the nomenclature of the enzyme cocktail to Cellic CTec3 HS, emphasizing that it is a commercial formulation composed of cellulases and hemicellulases. Regarding the correct commercial name of the enzyme cocktail used in this study, it is indeed Cellic® CTec3 HS (Novozymes), which is an enzyme mixture containing cellulase and hemicellulase activities. The manuscript was revised and updated in the Materials and Methods section (Section 2.9, Enzymatic Hydrolysis).
Comments 5. "It would be good to provide the mass of the biomass after hydrothermal and organic solvent extraction. Also, representing the results in terms of g in Table 1 would be suitable for better understanding, instead of %."
Response 5. We appreciate the suggestion, but we will keep the data as percentages only to avoid confusion for the reader. In pretreatment studies, it is already common practice to report results as percentages, and this does not pose any difficulties for readers due to the field's history of adopting this methodology. Anyone can easily convert to mass by taking a 100g aliquot and calculating the content of the different components based on the percentages. The reader can also use a 100g aliquot to determine the mass relative to the yields.
Comments 6. The sentence in line 273 is not clear.
Response 6. The sentence was changed in the manuscript for better understanding.
Commnets 7. Line 283: Mass loss of 8.79% in SBH? In Table 1, hydrothermal pretreatment greatly removed hemicellulose. Please check this data.
Response 7. We thank the reviewer for the observation, and indeed there was an error in the yield data that resulted in this low mass loss value. The yield was 71.21%, not 91.21%. This implies that the mass loss was 28.79%, not 8.79%. This data was corrected in the manuscript.
Comments 8. Figures 2 and 3 are redundant, as the composition data before and after pretreatment are already shown in Table 1. Reviewers and general readers can easily see the differences in cellulose, hemicellulose, and lignin before and after the pretreatment
Response 8. Thank you for your comment. Indeed, the information contained in Figures 2 and 3 is largely redundant with Table 1. We have accepted the suggestion to remove these figures to avoid repetition and simplify the presentation of the data.
Commnets 9. Lines 302 to 304: To understand the delignification of SB and SS from the selected pretreatments, it is essential to show the mass recovery after the pretreatment. For example, after hydrothermal pretreatment, 8.5 g of biomass was recovered from 10 g of pretreatment. In 8.5 g, what percentage of cellulose, hemicellulose, and lignin should be provided in the table for ease of understanding?”
Response 9. We appreciate the suggestion, but we will keep the data as percentages only to avoid confusion for the reader. In pretreatment studies, it is already common practice to report results as percentages, and this does not pose any difficulties for readers due to the field's history of adopting this methodology. Anyone can easily convert to mass by taking a 100g aliquot and calculating the content of the different components based on the percentages. The reader can also use a 100g aliquot to determine the mass relative to the yields.
Commnets 10. Lines 389 to 394: Several studies on hydrothermal pretreated biomass use cellulase and hemicellulase together or hemicellulose alone, achieving more than 80% hemicellulose hydrolysis. For the authors' information, after the hydrothermal pretreatment, enzymatic hydrolysis was conducted without separating the liquor, reaching more than 80% hemicellulose hydrolysis. Why did the authors state an adverse effect on hemicellulose due to the presence of lignin, etc.? Please check the literature. Has the author used the wrong enzyme for hemicellulose hydrolysis?
Response 10. We thank the reviewer for the observation and the reference to the literature. Indeed, several studies report high hemicellulose hydrolysis rates, exceeding 80%, when enzymatic hydrolysis is performed directly on pretreated biomass, without liquor separation. However, in the present study, we opted for a different approach: hydrolysis was performed specifically on hemicellulosic liquor, after pulp separation and the lignin precipitation step. This is because we conducted studies only with liquor, to generate input for pulp production companies that do not intend to use the pretreated pulp for 2G ethanol production, but could make better use of the liquor instead of simply burning it.
The cocktail used (Cellic® CTec3 HS, Novozymes) is a commercial formulation containing both cellulase and hemicellulase activities, but was developed primarily for the saccharification of cellulosic pulps. This may explain the lower efficiency observed in the hydrolysis of hemicelluloses in the isolated liquor. Furthermore, we believe that the residual presence of lignin and degradation compounds in the liquor may have exerted an inhibitory effect on the enzymes, which is consistent with other reports in the literature.
We recognize that the use of enzyme preparations more specifically targeted at hemicelluloses could increase the efficiency of the process, and this point was added to the Discussion section of the manuscript. However, the use of enzymes specific to hemicellulose hydrolysates should be explored in future work.
Comments 11. There are several articles published on the same topic that are more advanced than this work, including hydrothermal pretreatment without separating the liquors, using cellulase and hemicellulase for hydrolysis of cellulose and hemicellulose with a lower concentration of furfural and 5-HMF.
Response 11. We thank the reviewer for the observation and the reference to the literature. Indeed, several studies report high hemicellulose hydrolysis rates, exceeding 80%, when enzymatic hydrolysis is performed directly on pretreated biomass, without liquor separation. However, in the present study, we opted for a different approach: hydrolysis was performed specifically on hemicellulosic liquor, after pulp separation and the lignin precipitation step. This is because we conducted studies only with liquor, to generate input for pulp production companies that do not intend to use the pretreated pulp for 2G ethanol production, but could make better use of the liquor instead of simply burning it.
Comments 12. The authors provided a significant elevation of the study in the introduction, but nothing new in the materials and methods, results, and discussion.
Response 12. See the answer to the previous question.
Reviewer 2 Report
Comments and Suggestions for Authors
Reviewer Comments
This manuscript presents a straightforward study on organosolv and hydrothermal pretreatments of sugarcane bagasse and straw, followed by enzymatic hydrolysis of the hemicellulosic liquors for potential bioethanol production. The work is methodologically sound and aligns with current interests in lignocellulosic biomass valorization, demonstrating effective delignification and sugar release. However, the study feels somewhat conventional, with limited datasets that hinder robust conclusions and a lack of in-depth mechanistic analysis, which reduces its novelty and impact. While the topic is relevant, major revisions are needed to expand the experimental scope, provide more comprehensive data, and deepen the discussion of underlying mechanisms to elevate the manuscript to publication standards.
Major Suggestions: These points require significant revisions, including additional experiments, data collection, and expanded analysis to address the gaps in data quantity and mechanistic depth.
- Insufficient Data Quantity and Reproducibility:
The results section provides some quantitative data including compositional analysis in Table 1, line 258; TRS quantification in Table 3, line 370; and HPLC results in Table 4, line 408, but the datasets are notably sparse. For instance, while experiments are mentioned as being conducted in triplicate (e.g., line 146 for chemical composition and line 242 for enzymatic hydrolysis), the reported standard deviations are sometimes high (e.g., lignin content in SBH at 33.63 ± 6.28%, line 258),
suggesting variability that is not adequately addressed or explored. There is no inclusion of raw data, statistical analysis (e.g., ANOVA or t-tests to confirm significance), or additional replicates under varied conditions (e.g., different enzyme loadings or pretreatment times). This limits reliability, such as the "considerable increase" in sugars post-hydrolysis (line 379). To strengthen this, the authors should conduct and report at least 2-3 additional experimental runs, including error propagation analysis, and present supplementary figures (e.g., time-course data for hydrolysis yields).
- Limited Mechanistic Depth:
The discussion of pretreatment mechanisms and enzymatic hydrolysis is descriptive but lacks depth, often relying on general citations without integrating study-specific insights.
Lines 299-309: The increase in cellulose content and delignification is noted, but there is no detailed explanation of how organosolv vs. hydrothermal processes differentially affect hemicellulose solubilization or lignin redeposition at a molecular level (e.g., via changes in hydrogen bonding or acetyl group removal).
Lines 310-349: The SEM and XRD analyses are promising but underexplored. For example, no quantification of porosity changes or correlation with hydrolysis yields. The authors should expand this section with mechanistic models (e.g., discussing autohydrolysis in hydrothermal pretreatment via acetic acid release from hemicellulose) and integrate computational or spectroscopic data (e.g., FTIR for functional group changes) to provide a more profound understanding. This would address the conventional nature of the work and highlight novel contributions.
Lines 383-394: The challenges in hemicellulose hydrolysis are mentioned (e.g., heterogeneous structure and inhibitors), but the manuscript does not delve into why certain inhibitors (e.g., furfural in SBO at 20.714 g/L, line 408) form disproportionately or how they interact with the Cellic CTec3 enzyme cocktail.
Minor Suggestions: These are smaller issues that can be addressed through textual revisions and clarifications.
Line 117: The organosolv pretreatment pressure is given as "200 psi or 1.38 MPa" - specify if this is gauge or absolute pressure for reproducibility.
Line 236: The enzyme "Cellic CTec3" is mentioned without detailing its specific activity (e.g., FPU/mL) or source batch, which is important for replication.
Line 246: Minor grammatical issues appear, such as "ART dosage" (likely meant "TRS dosage," ).
Line 262: The comparison to Costa et al. (2021) is good, but briefly explain discrepancies (e.g., hemicellulose variations) rather than dismissing them as "cultivation conditions".
Lines 294 and 296: Fig. 2-3 effectively compare compositions but lack error bars, making it hard to assess variability.
Line 304: To maintain consistency with the data in Table 1, the phrase “39.16% lignin extraction” in this line should be changed to “39.16% delignification”.
Line 401: Based on the data in Table 4, the glucose content in the SBO, SSO, and SSH liquors did not increase after enzymatic hydrolysis; instead, it decreased. Additionally, the cellobiose content in the SSH liquor also decreased after enzymatic hydrolysis. Please check the data and provide analysis and discussion of the underlying principles.
Line 408: In Table 4, some values (e.g., arabinose at 0.000 ± 0.000 g/L) could be marked as "ND" (not detected) for clarity. The reference list is adequate but could be updated with more recent works (e.g., post-2023 studies on inhibitor mitigation in hemicellulosic liquors).
Lines 415-444: The conclusions repeat results without emphasizing implications for bioethanol scalability.
Line 431: Phrasing like "ART method" (clarify as "TRS method").
Author Response
Comments 1. This manuscript presents a straightforward study on organosolv and hydrothermal pretreatments of sugarcane bagasse and straw, followed by enzymatic hydrolysis of the hemicellulosic liquors for potential bioethanol production. The work is methodologically sound and aligns with current interests in lignocellulosic biomass valorization, demonstrating effective delignification and sugar release. However, the study feels somewhat conventional, with limited datasets that hinder robust conclusions and a lack of in-depth mechanistic analysis, which reduces its novelty and impact. While the topic is relevant, major revisions are needed to expand the experimental scope, provide more comprehensive data, and deepen the discussion of underlying mechanisms to elevate the manuscript to publication standards.
Response 1. We appreciate the reviewer's comments. We made several corrections to the manuscript to improve the quality of the work.
Comments 2. Major Suggestions: These points require significant revisions, including additional experiments, data collection, and expanded analysis to address the gaps in data quantity and mechanistic depth.
Response 2. We made several corrections to the manuscript to improve the quality of the work.
Commnets 3. Insufficient Data Quantity and Reproducibility:
The results section provides some quantitative data including compositional analysis in Table 1, line 258; TRS quantification in Table 3, line 370; and HPLC results in Table 4, line 408, but the datasets are notably sparse. For instance, while experiments are mentioned as being conducted in triplicate (e.g., line 146 for chemical composition and line 242 for enzymatic hydrolysis), the reported standard deviations are sometimes high (e.g., lignin content in SBH at 33.63 ± 6.28%, line 258), suggesting variability that is not adequately addressed or explored. There is no inclusion of raw data, statistical analysis (e.g., ANOVA or t-tests to confirm significance), or additional replicates under varied conditions (e.g., different enzyme loadings or pretreatment times). This limits reliability, such as the "considerable increase" in sugars post-hydrolysis (line 379). To strengthen this, the authors should conduct and report at least 2-3 additional experimental runs, including error propagation analysis, and present supplementary figures (e.g., time-course data for hydrolysis yields).
Response 3. We appreciate the observation. Indeed, a more extensive data set and additional statistical analyses could strengthen the robustness of the results. However, all experiments were performed in triplicate, and the values presented already include means and standard deviations. The observed variations, such as in the lignin content of SBH (33.63 ± 6.28%), reflect the natural heterogeneity of lignocellulosic biomass, which has a complex and not entirely uniform composition, even after homogenization.
Despite these variations, the main trends observed remained consistent across replicates, particularly the increase in sugar concentration after hydrolysis in the hydrothermal pretreatments compared to organosolv. We acknowledge that the lack of additional statistical tests is a limitation of the study, but we emphasize that the data obtained in triplicate were sufficient to demonstrate the trends discussed.
Comments 4. Limited Mechanistic Depth:
The discussion of pretreatment mechanisms and enzymatic hydrolysis is descriptive but lacks depth, often relying on general citations without integrating study-specific insights.
Lines 299-309: The increase in cellulose content and delignification is noted, but there is no detailed explanation of how organosolv vs. hydrothermal processes differentially affect hemicellulose solubilization or lignin redeposition at a molecular level (e.g., via changes in hydrogen bonding or acetyl group removal).
Lines 310-349: The SEM and XRD analyses are promising but underexplored. For example, no quantification of porosity changes or correlation with hydrolysis yields. The authors should expand this section with mechanistic models (e.g., discussing autohydrolysis in hydrothermal pretreatment via acetic acid release from hemicellulose) and integrate computational or spectroscopic data (e.g., FTIR for functional group changes) to provide a more profound understanding. This would address the conventional nature of the work and highlight novel contributions.
Lines 383-394: The challenges in hemicellulose hydrolysis are mentioned (e.g., heterogeneous structure and inhibitors), but the manuscript does not delve into why certain inhibitors (e.g., furfural in SBO at 20.714 g/L, line 408) form disproportionately or how they interact with the Cellic CTec3 enzyme cocktail.
Response 4. We made several adjustments to the manuscript to meet the reviewer's needs. Regarding the organosolv and hydrothermal processes, these are well-established in the literature and require no in-depth analysis in a scientific article aimed at researchers in the field, as everyone is familiar with the differences between these processes and their impact on the resulting products (pulp and liquor). Additionally, the main focus of this work is the hydrolysis and production of fermentable sugars, primarily from the liquor fraction, which is typically used only for combustion and is underutilized. This was an initial study to understand the processes involved in the use of liqueurs, which in this case were generated in acid-catalyzed pretreatments, which tend to generate higher amounts of inhibitors due to the dehydration of C5 and C6 sugars. Further studies will be conducted comparing these with liqueurs obtained through alkaline processes, in which inhibitor levels are much lower or nonexistent. Thus, a more in-depth discussion will be conducted and published in the future.
Comments 5. Minor Suggestions: These are smaller issues that can be addressed through textual revisions and clarifications.
Line 117: The organosolv pretreatment pressure is given as "200 psi or 1.38 MPa" - specify if this is gauge or absolute pressure for reproducibility.
Line 236: The enzyme "Cellic CTec3" is mentioned without detailing its specific activity (e.g., FPU/mL) or source batch, which is important for replication.
Line 246: Minor grammatical issues appear, such as "ART dosage" (likely meant "TRS dosage," ).
Line 262: The comparison to Costa et al. (2021) is good, but briefly explain discrepancies (e.g., hemicellulose variations) rather than dismissing them as "cultivation conditions".
Lines 294 and 296: Fig. 2-3 effectively compare compositions but lack error bars, making it hard to assess variability.
Line 304: To maintain consistency with the data in Table 1, the phrase “39.16% lignin extraction” in this line should be changed to “39.16% delignification”.
Line 401: Based on the data in Table 4, the glucose content in the SBO, SSO, and SSH liquors did not increase after enzymatic hydrolysis; instead, it decreased. Additionally, the cellobiose content in the SSH liquor also decreased after enzymatic hydrolysis. Please check the data and provide analysis and discussion of the underlying principles.
Line 408: In Table 4, some values (e.g., arabinose at 0.000 ± 0.000 g/L) could be marked as "ND" (not detected) for clarity. The reference list is adequate but could be updated with more recent works (e.g., post-2023 studies on inhibitor mitigation in hemicellulosic liquors).
Lines 415-444: The conclusions repeat results without emphasizing implications for bioethanol scalability.
Line 431: Phrasing like "ART method" (clarify as "TRS method").
Response 5. The points highlighted by the reviewer were corrected in the final manuscript. A general review of the text was carried out.
Reviewer 3 Report
Comments and Suggestions for Authors
The manuscript, titled “Organosolv and Hydrothermal Pretreatments of Sugarcane Bagasse and Straw, and Enzymatic Hydrolysis of Hemicellulosic Liquor for Bioethanol Production”, presents an interesting approach to extracting hemicellulose from biomass and conducting enzymatic hydrolysis for potential bioethanol production. This study is of high interest for green and sustainable production. However, several points need to be addressed before the manuscript can be considered for publication. Comments for the authors are as follows:
- In the Abstract, the author states, “Total reducing sugar concentrations increased significantly”; however, according to the results, this increase was not significant for the organosolv treatment, whereas it was significant for the hydrothermal one. This point should be revised and clarified.
- The Introduction is well-written and organized into a coherent narrative. However, it is suggested that in the last paragraph, the author use the present tense rather than the past tense (e.g., “The objective of the study is…”or “The specific objectives of this project are…”).
- In Section 2.1, the duration of the milling should be mentioned.
- In Line 117, the authors do not need to mention pressure using two different units; using only one unit is sufficient.
- In Line 119, the authors mention that for filtration, they used only water for washing. Please clarify whether an organic solvent (e.g., acetone) was also needed to wash any oil that may have been captured by the solid phase.
- Section 3.1.1is titled "SEM Analysis", but the authors also discuss the XRD results. However, the XRD patterns are not included in the manuscript and should be added.
- It is suggested that the authors, based on their obtained results, provide a deeper conclusion regarding which of the two applied pretreatment techniques is more promising and why.
Author Response
Comments 1. The manuscript, titled “Organosolv and Hydrothermal Pretreatments of Sugarcane Bagasse and Straw, and Enzymatic Hydrolysis of Hemicellulosic Liquor for Bioethanol Production”, presents an interesting approach to extracting hemicellulose from biomass and conducting enzymatic hydrolysis for potential bioethanol production. This study is of high interest for green and sustainable production. However, several points need to be addressed before the manuscript can be considered for publication. Comments for the authors are as follows:
Response 1. We thank the reviewer for recognizing the importance of the study. We also appreciate the suggestions for corrections.
Comments 2. In the Abstract, the author states, “Total reducing sugar concentrations increased significantly”; however, according to the results, this increase was not significant for the organosolv treatment, whereas it was significant for the hydrothermal one. This point should be revised and clarified.
Response 2. The manuscript was adjusted by removing the word “significantly”, especially because the measurement of what is significant can be incongruent.
Comments 3. The Introduction is well-written and organized into a coherent narrative. However, it is suggested that in the last paragraph, the author use the present tense rather than the past tense (e.g., “The objective of the study is…”or “The specific objectives of this project are…”).
Response 3. The sentence was rewritten as suggested.
Comments 4. In Section 2.1, the duration of the milling should be mentioned.
Response 4. The sentence was rewritten inserting the milling time.
Comments 5. In Line 117, the authors do not need to mention pressure using two different units; using only one unit is sufficient.
Response 5. The sentence was rewritten as suggested.
Comments 6. In Line 119, the authors mention that for filtration, they used only water for washing. Please clarify whether an organic solvent (e.g., acetone) was also needed to wash any oil that may have been captured by the solid phase.
Response 6. No other organic solvent was used to wash the pulp to avoid additional reactions and degradation of sugars since, after pretreatment, the lignocellulosic matrix is broken and the addition of another solvent can impact the quantification of sugars and measurement of pretreatment efficiency.
Comments 7. Section 3.1.1is titled "SEM Analysis", but the authors also discuss the XRD results. However, the XRD patterns are not included in the manuscript and should be added.
Response 7. We've adjusted the section titles, separating the SEM and XRD discussions into separate sections. We chose not to include the diffractogram image, but we appreciate the suggestion.
Comments 8. It is suggested that the authors, based on their obtained results, provide a deeper conclusion regarding which of the two applied pretreatment techniques is more promising and why.
Response 8. Corrections to the text of the conclusion were made.
Round 2
Reviewer 1 Report
Comments and Suggestions for Authors
The authors have addressed all the comments. The manuscript can be acceptable for publication.
Reviewer 2 Report
Comments and Suggestions for Authors
The authors did not response to the Reviewer's comments point by point.
Reviewer 3 Report
Comments and Suggestions for Authors
The modifications are satisfactory and the manuscript can be considered for publication.

