Obsidian at the Maddalena di Muccia (Marche, Central Italy) Archaeological Site: Evidence for Three Volcanic Sources in a Multi-Phase Context

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The manuscript submitted by Acquafredda et al. would fit well within the scope of Heritage. The paper is clearly written and shows no major structural weaknesses. It is furthermore well illustrated with an appropriate set of tables and figures and is supported by an up-to-date bibliography. The research team is well recognised for its work on obsidian sourcing and enjoys strong international visibility.

The methodology used by Acquafredda et al. for obsidian sourcing, based on a previously published geochemical/mineralogical coupling, has been thoroughly validated and proves to be particularly robust (see references).

The approach presented in the manuscript satisfactorily integrates archaeological research questions.

If I were to make a single remark, the authors could consider expanding the brief literature review on obsidian sourcing (lines 48–53).

I recommend the publication of this manuscript.

Author Response

 

1. The reviewer 7 wrote: The manuscript submitted by Acquafredda et al. would fit well within the scope of Heritage. The paper is clearly written and shows no major structural weaknesses. It is furthermore well illustrated with an appropriate set of tables and figures and is supported by an up-to-date bibliography. The research team is well recognised for its work on obsidian sourcing and enjoys strong international visibility.

 

The methodology used by Acquafredda et al. for obsidian sourcing, based on a previously published geochemical/mineralogical coupling, has been thoroughly validated and proves to be particularly robust (see references).

 

The approach presented in the manuscript satisfactorily integrates archaeological research questions.

Our replay is: We thank the reviewer for appreciating all this aspects of our manuscript.

 

2. The reviewer 7 wrote: If I were to make a single remark, the authors could consider expanding the brief literature review on obsidian sourcing (lines 48–53).

Our replay is: We agree with the reviewer that a more detailed discussion of the bibliography would have enriched the paper, but the various techniques have already been discussed in the bibliography cited [9-27].

 

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The article presents the results of provenance studies of obsidian artefacts recovered from Early Neolithic to Chalcolithic (2900-2200 BC) contexts at the site of Maddalena di Muccia in the Umbria-Marche Apennines of (Central) Italy. The site has produced a wealth of evidence about early farming communities in the region, offering insights into their economy, environment and regional cultural connections.

This article builds on previous archaeological and palaeoenvironmental research at the site by reporting the results of geochemical and petrographic analyses of 11 obsidian artefacts (13 pieces) — 4 from Early Neolithic contexts, 6 from well-documented or presumed Chalcolithic contexts, and one “undated” find.

While I think this is a good paper that makes a significant contribution to obsidian provenance studies in the Central Mediterranean region, there is room for improvement, as follows:

A. The use of English, including grammar and spelling, needs to be improved in the interests of clarity and accuracy. 

B. There is some key information that is missing from the text, which really ought to be there. For example:

1. What is the total number of obsidian pieces recovered in the excavations?
2. How many/what proportion have been subjected to geochemical characterisation?
3. Were any obsidian artefacts recovered from Middle Neolithic contexts?
4. Why was the obsidian artefact from the Late Neolithic pit (SU12) not analysed?
5. How many ¹⁴C dates have been obtained from the site and from features that contained obsidian?
6. Are the radiocarbon ages cited in the paper quoted in cal BC years?
7. In the paper, age ranges (presumably calibrated) are reported in years BC and BCE – what is the difference? If none, please cite dates in only one format.
8. In Table 1, what does the term “Debris” mean – fragment, debitage?
9. Numerous studies by Robert Tykot and others have shown that the mid-Z trace elements (Rb, Sr, Y, Zr and Nb) measured by pXRF can distinguish between most/all known Central Mediterranean obsidian sources and sub-sources. What, then, is the benefit of using high-end (slower and more expensive) laboratory-based methods like WD-XRF and SEM-EDS? Some discussion of this issue would be useful.

Comments for author File: Comments.pdf

Comments on the Quality of English Language

There is substantial room for improvement here. I have edited/annotated the text to indicate the main changes that are needed. One issue is inconsistent spelling — I suspect that various authors have contributed to the text, some using American conventions, others British conventions. I have opted for British spellings for the main text. But note that the Bibliography (article or book titles) should follow the conventions on spelling and capitalisation used in the original publications.

Author Response

1. The reviewer 2 wrote: Line 38: at the end of the sentence With a research history spanning more than fifty years [1-8], obsidian provenance studies have long contributed substantially to reconstructing prehistoric interaction networks across the Mediterranean, please add the following reference:
2. J. Ibáñez, D. Ortega, D. Campos, L. Khalidi, V. Méndez, 2015. Testing complex networks of interaction at the onset of the Near Eastern Neolithic using modelling of obsidian exchange. Journal of the Royal Society Interface 12, 20150210.10.1098/rsif.2015.0210

Our replay is: In this point of the manuscript, we have cited all the research concerning the characterization and provenance of obsidian, with particular reference to analytical techniques: it does not appear that the paper suggested by reviewer 2 concerns the characterization of obsidian and, moreover, it refers to a mathematical modeling of obsidian exchange in the Near Eastern Neolithic: sorry, but we disagree with reviewer 2 on adding the reference.

 

2. The reviewer 2 wrote: The text in the line 40-43, Therefore, the increasing availability of portable instruments combined with non-invasive analytical methods provides an opportunity to significantly broaden the range of analysable artefacts and to generate more comprehensive and comparable datasets, should be referenced too. Or, the authors need to rephrase the sentence, since they didn’t use the portable instruments in the presented study.

Our replay is: We have changed the sentence at lines 40-43

 

3. The reviewer 2 wrote: In the line 55, the acronym EDS should be explained. If this acronym refers to scanning electron microscopy coupled with energy dispersive spectrometer (SEM-EDS), as mentioned in line 51, that should be properly written. If not, please refer properly to avoid confusing the reader.

Our replay is: We have corrected the sentence “ED XRF” with “Energy-dispersive XRF”

 

4. The reviewer 2 wrote: The text in lines 54-60 should be merged so the reference is meaningful.

Our replay is: At lines 53-54 we have cut the sentence: “and Energy Dispersive (ED) Spectrometers [14,15].”

 

5. The reviewer 2 wrote: The text in lines 61-64 seems unfinished. Please rephrase or avoid.

Our replay is: The authors find the text clear: the same concepts has been written, with other words, in previous papers published in international journals.

 

6. The reviewer 2 wrote: There is a typo in line 71.

Our replay is: We have corrected the typo

 

7. The reviewer 2 wrote: In the caption of Figure 1, please add an explanation for the green and black text.

Our replay is: We have changed the figure 1

 

8. The reviewer 2 wrote: In line 80, correct Impressa to Impresso (or Cardium) pottery as it is in English.

Our replay is: The term ‘Impresso pottery’ is incorrect, as it constitutes an improper distortion of the correct Italian term ‘Impressa’. We therefore changed it to ‘impressed’.

 

9. The reviewer 2 wrote: In line 86 there is one more typo in the caption of Table 1.

Our replay is: We have corrected the typo

 

10. The reviewer 2 wrote: Please check carefully in Table 1 what should be written: SU (SU- stratigraphic unit, mentioned above in the text) or US.

Our replay is: We have corrected ‘SU’

 

11. The reviewer 2 wrote: After Table1 the photo of the obsidian samples should be provided. Together with the scale.

Our replay is: In Table 1, we have included photos of the analysed obsidian samples, together with a scale bar.

 

12. The reviewer 2 wrote: The statement completely non-destructive analytical procedures in line 132 should be reconsidered, especially for the WD-XRF analytical technique, according to the obsidian dimensions. The term “non- destructive” may be used correctly, but please provide more information on the sample preparation for WD-XRF analysis.

Our replay is: We strongly disagree with the reviewer's opinion: WD-XRF analyses on obsidian, using the X-ray intensity ratio procedure, by many authors are correctly considered to be completely non-destructive, as already explained in many previous research, including the one cited in the text (Acquafredda, et al. 2018 “Reassessment of the WD-XRF method for obsidian provenance shareable databases”. Quartern. Int. and the reference therein).

Anyway, at the beginning of the “Results” section we have changed the sentence “All obsidian samples were characterised using non-destructive WD-XRF analysis, which allows information on their source area to be obtained very quickly, in less than 30 minutes per sample [31], by measuring the net intensities of some trace elements, namely Rb, Sr, Y, Zr, and Nb (Table 2).” with the sentence:

“Before discussing the results, it is worth briefly describing the procedure used at the University of Bari Aldo Moro, which allows very accurate analytical data, to be obtained with a completely non-destructive techniques, useful for attributing the obsidian source areas.

An initial characterization is performed using WD-XRF, which allows the sample to be characterized very quickly, in 30 minutes [31], without having to treat the sample in any way other than washing it in a beaker containing distilled water. In this way, the glass and microphenocrysts are analysed, and the results are comparable only with laboratories that have used the same technique to measure the X-ray intensities of trace elements such as Rb, Sr, Y, Zr, and Nb.

If the results obtained leave room for doubt, further analyses are carried out using SEM-EDS, which allows both the analysis of the major elements of the glass, if it is unaltered, and that of the microphenocrysts [2, 11, 32]. The analysis of microphenocrysts certify the datum because it is not affected by glass alteration phenomena: the artifacts are normally subjected to alteration burial processes, due to waters circulating within sediments [38]. These alteration processes can affect alkaline elements mobility, above all Na, and the formation of thin carbonate film incrustations [31, 38 and the references therein], sometimes even enlarging the uranium-238 fission tracks [32]. The obsidian surface alteration can in some way interfere with WD-XRF analysis even making it impossible to analyse by SEM-EDS the glass.

The precision and accuracy of analytical data were extensively discussed for the first time in 1999 [2] and in other successive papers [11, 31, 32].

All obsidians from the Maddalena di Muccia site were previously characterised using WD-XRF analysis, by measuring the net intensities of some trace elements, namely Rb, Sr, Y, Zr, and Nb [31 (Table 2).

 

13. The reviewer 2 wrote: The measurement time per sample using WD-XRF spectrometry should be stated, for example, in lines 153-156.

Our replay is: In the manuscript we have written “The Rb, Sr, Y, Zr and Nb X-ray emission lines were measured according to the procedure of [31] and [33,34]”: in the papers 31, 33 and 34 are reported, in detail, the procedures, also the measurement time; the measurement time is also indicated at lines 188-189 of the PDF “obtained very quickly, in less than 30 minutes per sample [31]”

 

14. The reviewer 2 wrote: In line 187, the text is partially meaningless. By WD-XRF spectrometry, we can not gain information about the source of the obsidian. This needs to be corrected.

Our replay is: We strongly disagree with the reviewer's opinion: WD-XRF analyses on obsidian, using the X-ray intensity ratio procedure, by many authors are considered one of the most accurate possibility to have obsidian source area information that are also shareable (Acquafredda, et al. 2018 “Reassessment of the WD-XRF method for obsidian provenance shareable databases”. Quarternary International, and the reference therein).

 

15. The reviewer 2 wrote: In line 205, there is one more typo, 19b, should be 10b.

Our replay is: We have corrected the typo

 

16. The reviewer 2 wrote: I suggest rephrasing the text: The peculiarity that three different sources area are present in the same archaeological site, one of which even represented by two sub-sources, suggested that the analytical data obtained using WD-XRF should be further constrained using SEM-EDS investigations (lines 207-210). As it is written now, one can conclude that the WD-XRF results, compared with previous results, are incorrect for some reason (analytical procedure, archaeological context, etc.).

Our replay is: That's exactly right. Often, the degree of alteration of obsidian samples does not allow for a definite attribution, so it is necessary to use rapid analysis via WD-XRF of the total sample, to which glass analysis (SEM-EDS) may also be added, and in some special cases, microphenocrysts analysis.

Anyway, at the beginning of the “Results” section we have changed the sentence “All obsidian samples were characterised using non-destructive WD-XRF analysis, which allows information on their source area to be obtained very quickly, in less than 30 minutes per sample [31], by measuring the net intensities of some trace elements, namely Rb, Sr, Y, Zr, and Nb (Table 2).” with the sentence:

“Before discussing the results, it is worth briefly describing the procedure used at the University of Bari Aldo Moro, which allows very accurate analytical data, to be obtained with a completely non-destructive techniques, useful for attributing the obsidian source areas.

An initial characterization is performed using WD-XRF, which allows the sample to be characterized very quickly, in 30 minutes [31], without having to treat the sample in any way other than washing it in a beaker containing distilled water. In this way, the glass and microphenocrysts are analysed, and the results are comparable only with laboratories that have used the same technique to measure the X-ray intensities of trace elements such as Rb, Sr, Y, Zr, and Nb.

If the results obtained leave room for doubt, further analyses are carried out using SEM-EDS, which allows both the analysis of the major elements of the glass, if it is unaltered, and that of the microphenocrysts [2, 11, 32]. The analysis of microphenocrysts certify the datum because it is not affected by glass alteration phenomena: the artifacts are normally subjected to alteration burial processes, due to waters circulating within sediments [38]. These alteration processes can affect alkaline elements mobility, above all Na, and the formation of thin carbonate film incrustations [31, 38 and the references therein], sometimes even enlarging the uranium-238 fission tracks [32]. The obsidian surface alteration can in some way interfere with WD-XRF analysis even making it impossible to analyse by SEM-EDS the glass.

The precision and accuracy of analytical data were extensively discussed for the first time in 1999 [2] and in other successive papers [11, 31, 32].

All obsidians from the Maddalena di Muccia site were previously characterised using WD-XRF analysis, by measuring the net intensities of some trace elements, namely Rb, Sr, Y, Zr, and Nb [31 (Table 2).

 

17. The reviewer 2 wrote: The results in Tables 3, 4, and 5 should be presented with the measurement uncertainty.

Our replay is: At the beginning of the “Results” section we have changed the sentence “All obsidian samples were characterised using non-destructive WD-XRF analysis, which allows information on their source area to be obtained very quickly, in less than 30 minutes per sample [31], by measuring the net intensities of some trace elements, namely Rb, Sr, Y, Zr, and Nb (Table 2).” with the sentence:

“Before discussing the results, it is worth briefly describing the procedure used at the University of Bari Aldo Moro, which allows very accurate analytical data, to be obtained with a completely non-destructive techniques, useful for attributing the obsidian source areas.

An initial characterization is performed using WD-XRF, which allows the sample to be characterized very quickly, in 30 minutes [31], without having to treat the sample in any way other than washing it in a beaker containing distilled water. In this way, the glass and microphenocrysts are analysed, and the results are comparable only with laboratories that have used the same technique to measure the X-ray intensities of trace elements such as Rb, Sr, Y, Zr, and Nb.

If the results obtained leave room for doubt, further analyses are carried out using SEM-EDS, which allows both the analysis of the major elements of the glass, if it is unaltered, and that of the microphenocrysts [2, 11, 32]. The analysis of microphenocrysts certify the datum because it is not affected by glass alteration phenomena: the artifacts are normally subjected to alteration burial processes, due to waters circulating within sediments [38]. These alteration processes can affect alkaline elements mobility, above all Na, and the formation of thin carbonate film incrustations [31, 38 and the references therein], sometimes even enlarging the uranium-238 fission tracks [32]. The obsidian surface alteration can in some way interfere with WD-XRF analysis even making it impossible to analyse by SEM-EDS the glass.

The precision and accuracy of analytical data were extensively discussed for the first time in 1999 [2] and in other successive papers [11, 31, 32].

All obsidians from the Maddalena di Muccia site were previously characterised using WD-XRF analysis, by measuring the net intensities of some trace elements, namely Rb, Sr, Y, Zr, and Nb [31 (Table 2).

 

18. The reviewer 2 wrote: The results described in the text: In some cases, however, glass analysis confirmed that samples 10b and 11 originated from Lipari and that samples 5, 6 and 9 originated from Monte Arci, and in addition attributing sub-source S.A. to samples 5 and 9 and sub-source S.C. to sample 6 (Figure 5). (lines 218-220), cannot be easily noticed in Figure.

Our replay is: The authors partly understand the reviewer's concerns: the uncertain attribution of samples 5, 6, and 9 in Figure 5 (the samples had already been attributed by looking at Figure 4) led to the decision to also analyse the microphenocrysts.

 

19. The reviewer 2 wrote: The association of the chemical composition of sample 6 with the subgroup Arci S.C. is questionable.

Our replay is: We strongly disagree with the reviewer's opinion: a quick look to figures 4 and 9 allows for the certain attribution of sample 6 to sub-source S.C.

 

20. The reviewer 2 wrote: For the text: Unfortunately, for the samples ascribed to Palmarola geological outcrops using WD-XRF, it was not possible to confirm their origin solely through glass analysis, both due to surface alteration of the artefact and the presence of thin carbonate concretions. (lines 227-229) is necessary to indicate from where it can be seen.

Our replay is: Immediately after the sentence reported here by the Reviewer 2 (…. Concretions) in the text of the manuscript is written the explanation: “Fortunately, the presence of pyroxene crystals (Figure 6), although very small and often with a skeletal texture, allowed the attribution of some samples to the Palmarola source area (Figure 7, Table 4).

 

21. The reviewer 2 wrote: In the caption of the tables 4 and 5, it is necessary to indicate the measure of the oxides composition.

Our replay is: We have added in the caption of table 3, 4 and 5 the sentence: “(oxides expressed in wt%)”.

 

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

All the comments are listed in the attached document

Comments for author File: Comments.pdf

Author Response

MDPI REVIEW REPORT 2
1. The reviewer 2 wrote: Line 38: at the end of the sentence With a research history spanning
more than fifty years [1-8], obsidian provenance studies have long contributed substantially to
reconstructing prehistoric interaction networks across the Mediterranean, please add the
following reference:
J. J. Ibáñez, D. Ortega, D. Campos, L. Khalidi, V. Méndez, 2015. Testing complex networks of
interaction at the onset of the Near Eastern Neolithic using modelling of obsidian exchange.
Journal of the Royal Society Interface 12, 20150210.10.1098/rsif.2015.0210
Our replay is: In this point of the manuscript, we have cited all the research concerning the
characterization and provenance of obsidian, with particular reference to analytical techniques: it
does not appear that the paper suggested by reviewer 2 concerns the characterization of obsidian
and, moreover, it refers to a mathematical modeling of obsidian exchange in the Near Eastern
Neolithic: sorry, but we disagree with reviewer 2 on adding the reference.
2. The reviewer 2 wrote: The text in the line 40-43, Therefore, the increasing availability of
portable instruments combined with non-invasive analytical methods provides an opportunity to
significantly broaden the range of analysable artefacts and to generate more comprehensive and
comparable datasets, should be referenced too. Or, the authors need to rephrase the sentence,
since they didn’t use the portable instruments in the presented study.
Our replay is: We have changed the sentence at lines 40-43
3. The reviewer 2 wrote: In the line 55, the acronym EDS should be explained. If this acronym
refers to scanning electron microscopy coupled with energy dispersive spectrometer (SEM-EDS),
as mentioned in line 51, that should be properly written. If not, please refer properly to avoid
confusing the reader.
Our replay is: We have corrected the sentence “ED XRF” with “Energy-dispersive XRF”
4. The reviewer 2 wrote: The text in lines 54-60 should be merged so the reference is meaningful.
Our replay is: At lines 53-54 we have cut the sentence: “and Energy Dispersive (ED) Spectrometers
[14,15].”
5. The reviewer 2 wrote: The text in lines 61-64 seems unfinished. Please rephrase or avoid.
Our replay is: The authors find the text clear: the same concepts has been written, with other
words, in previous papers published in international journals.
6. The reviewer 2 wrote: There is a typo in line 71.
Our replay is: We have corrected the typo
7. The reviewer 2 wrote: In the caption of Figure 1, please add an explanation for the green and
black text.
Our replay is: We have changed the figure 1
8. The reviewer 2 wrote: In line 80, correct Impressa to Impresso (or Cardium) pottery as it is in
English.
Our replay is: The term ‘Impresso pottery’ is incorrect, as it constitutes an improper distortion of
the correct Italian term ‘Impressa’. We therefore changed it to ‘impressed’.
9. The reviewer 2 wrote: In line 86 there is one more typo in the caption of Table 1.
Our replay is: We have corrected the typo
10. The reviewer 2 wrote: Please check carefully in Table 1 what should be written: SU (SUstratigraphic
unit, mentioned above in the text) or US.
Our replay is: We have corrected ‘SU’
11. The reviewer 2 wrote: After Table1 the photo of the obsidian samples should be provided.
Together with the scale.
Our replay is: In Table 1, we have included photos of the analysed obsidian samples, together with
a scale bar.
12. The reviewer 2 wrote: The statement completely non-destructive analytical procedures in line
132 should be reconsidered, especially for the WD-XRF analytical technique, according to the
obsidian dimensions. The term “non- destructive” may be used correctly, but please provide more
information on the sample preparation for WD-XRF analysis.
Our replay is: We strongly disagree with the reviewer's opinion: WD-XRF analyses on obsidian,
using the X-ray intensity ratio procedure, by many authors are correctly considered to be
completely non-destructive, as already explained in many previous research, including the one
cited in the text (Acquafredda, et al. 2018 “Reassessment of the WD-XRF method for obsidian
provenance shareable databases”. Quartern. Int. and the reference therein).
Anyway, at the beginning of the “Results” section we have changed the sentence “All obsidian
samples were characterised using non-destructive WD-XRF analysis, which allows information on
their source area to be obtained very quickly, in less than 30 minutes per sample [31], by
measuring the net intensities of some trace elements, namely Rb, Sr, Y, Zr, and Nb (Table 2).” with
the sentence:
“Before discussing the results, it is worth briefly describing the procedure used at the University of
Bari Aldo Moro, which allows very accurate analytical data, to be obtained with a completely nondestructive
techniques, useful for attributing the obsidian source areas.
An initial characterization is performed using WD-XRF, which allows the sample to be
characterized very quickly, in 30 minutes [31], without having to treat the sample in any way other
than washing it in a beaker containing distilled water. In this way, the glass and microphenocrysts
are analysed, and the results are comparable only with laboratories that have used the same
technique to measure the X-ray intensities of trace elements such as Rb, Sr, Y, Zr, and Nb.
If the results obtained leave room for doubt, further analyses are carried out using SEM-EDS,
which allows both the analysis of the major elements of the glass, if it is unaltered, and that of the
microphenocrysts [2, 11, 32]. The analysis of microphenocrysts certify the datum because it is not
affected by glass alteration phenomena: the artifacts are normally subjected to alteration burial
processes, due to waters circulating within sediments [38]. These alteration processes can affect
alkaline elements mobility, above all Na, and the formation of thin carbonate film incrustations
[31, 38 and the references therein], sometimes even enlarging the uranium-238 fission tracks [32].
The obsidian surface alteration can in some way interfere with WD-XRF analysis even making it
impossible to analyse by SEM-EDS the glass.
The precision and accuracy of analytical data were extensively discussed for the first time in 1999
[2] and in other successive papers [11, 31, 32].
All obsidians from the Maddalena di Muccia site were previously characterised using WD-XRF
analysis, by measuring the net intensities of some trace elements, namely Rb, Sr, Y, Zr, and Nb [31
(Table 2).
13. The reviewer 2 wrote: The measurement time per sample using WD-XRF spectrometry should
be stated, for example, in lines 153-156.
Our replay is: In the manuscript we have written “The Rb, Sr, Y, Zr and Nb X-ray emission lines
were measured according to the procedure of [31] and [33,34]”: in the papers 31, 33 and 34 are
reported, in detail, the procedures, also the measurement time; the measurement time is also
indicated at lines 188-189 of the PDF “obtained very quickly, in less than 30 minutes per sample
[31]”
14. The reviewer 2 wrote: In line 187, the text is partially meaningless. By WD-XRF spectrometry,
we can not gain information about the source of the obsidian. This needs to be corrected.
Our replay is: We strongly disagree with the reviewer's opinion: WD-XRF analyses on obsidian,
using the X-ray intensity ratio procedure, by many authors are considered one of the most
accurate possibility to have obsidian source area information that are also shareable
(Acquafredda, et al. 2018 “Reassessment of the WD-XRF method for obsidian provenance
shareable databases”. Quarternary International, and the reference therein).
15. The reviewer 2 wrote: In line 205, there is one more typo, 19b, should be 10b.
Our replay is: We have corrected the typo
16. The reviewer 2 wrote: I suggest rephrasing the text: The peculiarity that three different
sources area are present in the same archaeological site, one of which even represented by two
sub-sources, suggested that the analytical data obtained using WD-XRF should be further
constrained using SEM-EDS investigations (lines 207-210). As it is written now, one can conclude
that the WD-XRF results, compared with previous results, are incorrect for some reason (analytical
procedure, archaeological context, etc.).
Our replay is: That's exactly right. Often, the degree of alteration of obsidian samples does not
allow for a definite attribution, so it is necessary to use rapid analysis via WD-XRF of the total
sample, to which glass analysis (SEM-EDS) may also be added, and in some special cases,
microphenocrysts analysis.
Anyway, at the beginning of the “Results” section we have changed the sentence “All obsidian
samples were characterised using non-destructive WD-XRF analysis, which allows information on
their source area to be obtained very quickly, in less than 30 minutes per sample [31], by
measuring the net intensities of some trace elements, namely Rb, Sr, Y, Zr, and Nb (Table 2).” with
the sentence:
“Before discussing the results, it is worth briefly describing the procedure used at the University of
Bari Aldo Moro, which allows very accurate analytical data, to be obtained with a completely nondestructive
techniques, useful for attributing the obsidian source areas.
An initial characterization is performed using WD-XRF, which allows the sample to be
characterized very quickly, in 30 minutes [31], without having to treat the sample in any way other
than washing it in a beaker containing distilled water. In this way, the glass and microphenocrysts
are analysed, and the results are comparable only with laboratories that have used the same
technique to measure the X-ray intensities of trace elements such as Rb, Sr, Y, Zr, and Nb.
If the results obtained leave room for doubt, further analyses are carried out using SEM-EDS,
which allows both the analysis of the major elements of the glass, if it is unaltered, and that of the
microphenocrysts [2, 11, 32]. The analysis of microphenocrysts certify the datum because it is not
affected by glass alteration phenomena: the artifacts are normally subjected to alteration burial
processes, due to waters circulating within sediments [38]. These alteration processes can affect
alkaline elements mobility, above all Na, and the formation of thin carbonate film incrustations
[31, 38 and the references therein], sometimes even enlarging the uranium-238 fission tracks [32].
The obsidian surface alteration can in some way interfere with WD-XRF analysis even making it
impossible to analyse by SEM-EDS the glass.
The precision and accuracy of analytical data were extensively discussed for the first time in 1999
[2] and in other successive papers [11, 31, 32].
All obsidians from the Maddalena di Muccia site were previously characterised using WD-XRF
analysis, by measuring the net intensities of some trace elements, namely Rb, Sr, Y, Zr, and Nb [31
(Table 2).
17. The reviewer 2 wrote: The results in Tables 3, 4, and 5 should be presented with the
measurement uncertainty.
Our replay is: At the beginning of the “Results” section we have changed the sentence “All
obsidian samples were characterised using non-destructive WD-XRF analysis, which allows
information on their source area to be obtained very quickly, in less than 30 minutes per sample
[31], by measuring the net intensities of some trace elements, namely Rb, Sr, Y, Zr, and Nb (Table
2).” with the sentence:
“Before discussing the results, it is worth briefly describing the procedure used at the University of
Bari Aldo Moro, which allows very accurate analytical data, to be obtained with a completely nondestructive
techniques, useful for attributing the obsidian source areas.
An initial characterization is performed using WD-XRF, which allows the sample to be
characterized very quickly, in 30 minutes [31], without having to treat the sample in any way other
than washing it in a beaker containing distilled water. In this way, the glass and microphenocrysts
are analysed, and the results are comparable only with laboratories that have used the same
technique to measure the X-ray intensities of trace elements such as Rb, Sr, Y, Zr, and Nb.
If the results obtained leave room for doubt, further analyses are carried out using SEM-EDS,
which allows both the analysis of the major elements of the glass, if it is unaltered, and that of the
microphenocrysts [2, 11, 32]. The analysis of microphenocrysts certify the datum because it is not
affected by glass alteration phenomena: the artifacts are normally subjected to alteration burial
processes, due to waters circulating within sediments [38]. These alteration processes can affect
alkaline elements mobility, above all Na, and the formation of thin carbonate film incrustations
[31, 38 and the references therein], sometimes even enlarging the uranium-238 fission tracks [32].
The obsidian surface alteration can in some way interfere with WD-XRF analysis even making it
impossible to analyse by SEM-EDS the glass.
The precision and accuracy of analytical data were extensively discussed for the first time in 1999
[2] and in other successive papers [11, 31, 32].
All obsidians from the Maddalena di Muccia site were previously characterised using WD-XRF
analysis, by measuring the net intensities of some trace elements, namely Rb, Sr, Y, Zr, and Nb [31
(Table 2).
18. The reviewer 2 wrote: The results described in the text: In some cases, however, glass analysis
confirmed that samples 10b and 11 originated from Lipari and that samples 5, 6 and 9 originated
from Monte Arci, and in addition attributing sub-source S.A. to samples 5 and 9 and sub-source
S.C. to sample 6 (Figure 5). (lines 218-220), cannot be easily noticed in Figure.
Our replay is: The authors partly understand the reviewer's concerns: the uncertain attribution of
samples 5, 6, and 9 in Figure 5 (the samples had already been attributed by looking at Figure 4) led
to the decision to also analyse the microphenocrysts.
19. The reviewer 2 wrote: The association of the chemical composition of sample 6 with the
subgroup Arci S.C. is questionable.
Our replay is: We strongly disagree with the reviewer's opinion: a quick look to figures 4 and 9
allows for the certain attribution of sample 6 to sub-source S.C.
20. The reviewer 2 wrote: For the text: Unfortunately, for the samples ascribed to Palmarola
geological outcrops using WD-XRF, it was not possible to confirm their origin solely through glass
analysis, both due to surface alteration of the artefact and the presence of thin carbonate
concretions. (lines 227-229) is necessary to indicate from where it can be seen.
Our replay is: Immediately after the sentence reported here by the Reviewer 2 (…. Concretions) in
the text of the manuscript is written the explanation: “Fortunately, the presence of pyroxene
crystals (Figure 6), although very small and often with a skeletal texture, allowed the attribution of
some samples to the Palmarola source area (Figure 7, Table 4).
21. The reviewer 2 wrote: In the caption of the tables 4 and 5, it is necessary to indicate the measure
of the oxides composition.
Our replay is: We have added in the caption of table 3, 4 and 5 the sentence: “(oxides expressed in
wt%)”.

Author Response File: Author Response.pdf

Reviewer 4 Report

Comments and Suggestions for Authors

This manuscript presents a valuable petroarchaeometric study of obsidian artefacts from the Maddalena di Muccia site using non-destructive WD-XRF and SEM-EDS analyses. The analytical dataset is solid, and the integration of archaeological context with geochemical provenance determination represents a relevant contribution, particularly in documenting multiple obsidian sources, including Monte Arci sub-sources, within an Adriatic context.

However, the manuscript would benefit from substantial revision to improve clarity, structure, and analytical synthesis. Large parts of the Results and Discussion sections are descriptive, and the key outcomes of the study are not always clearly synthesized. The authors are encouraged to better highlight the main analytical patterns and to clarify how the presented data significantly refine current understanding of obsidian circulation networks. Stronger links between geochemical results and archaeological interpretation would also improve the overall impact of the study.

In addition, alteration effects on obsidian chemistry are acknowledged but not fully discussed in terms of their implications for provenance reliability. A clearer explanation of how analytical methods remain robust under such conditions would strengthen confidence in the results. The analytical workflow, particularly the complementary roles of WD-XRF and SEM-EDS, should also be clarified.

Finally, some sections could be streamlined to reduce repetition, figure discussions could be more explicitly integrated into the text, and minor language improvements would enhance readability.

Comments for author File: Comments.pdf

Comments on the Quality of English Language

The English language is generally understandable, but several sections contain long or complex sentences, minor grammatical issues, and occasional terminology inconsistencies. Careful language editing would improve clarity, readability, and overall flow of the manuscript, particularly in the Results and Discussion sections.

Author Response

1. MDPI REVIEW REPORT 3
   Major Comments
   1. The reviewer 3 wrote: Insufficient analytical synthesis in Results section (Lines 185–267): The
   Results section largely consists of presentation of numerical datasets and source assignments
   without sufficiently summarizing the analytical outcomes. Key patterns, such as chronological
   shifts in obsidian sources or differential distribution across contexts, are not synthesized
   quantitatively.
   Recommendation: Include summary statistics or diagrams summarizing source proportions by
   phase. Provide clearer statements indicating analytical confidence levels in provenance
   assignments.
   Our replay is: The data presented in Table 7 derive from studies carried out over an extended
   period of time and within different analytical and research frameworks; consequently, the
   proportion of characterised artefacts relative to the total assemblages cannot always be
   established, and the resulting datasets should not be regarded as statistically comparable.
   Nevertheless, when considered collectively, these data provide valuable evidence for
   reconstructing the spatial extent and diachronic development of the circulation of different
   obsidian sources over time.
   2. The reviewer 3 wrote: Provenance attribution relies on assumed robustness rather than
   demonstrated reliability (Lines 210–212; 227–232): The manuscript notes chemical alteration
   (especially Na depletion), yet does not quantitatively demonstrate how this affects provenance
   discrimination. It is unclear how robust WD-XRF trace element ratios remain under alteration.
   Recommendation: Provide justification or references demonstrating stability of chosen ratios.
   Explicitly discuss analytical uncertainty and error margins.
   Our replay is: At the beginning of the “Results” section we have changed the sentence “All
   obsidian samples were characterised using non-destructive WD-XRF analysis, which allows
   information on their source area to be obtained very quickly, in less than 30 minutes per sample
   [31], by measuring the net intensities of some trace elements, namely Rb, Sr, Y, Zr, and Nb (Table
   2).” with the sentence:
   “Before discussing the results, it is worth briefly describing the procedure used at the University of
   Bari Aldo Moro, which allows very accurate analytical data, to be obtained with a completely nondestructive
   techniques, useful for attributing the obsidian source areas.
   An initial characterization is performed using WD-XRF, which allows the sample to be
   characterized very quickly, in 30 minutes [31], without having to treat the sample in any way other
   than washing it in a beaker containing distilled water. In this way, the glass and microphenocrysts
   are analysed, and the results are comparable only with laboratories that have used the same
   technique to measure the X-ray intensities of trace elements such as Rb, Sr, Y, Zr, and Nb.
   If the results obtained leave room for doubt, further analyses are carried out using SEM-EDS,
   which allows both the analysis of the major elements of the glass, if it is unaltered, and that of the
   microphenocrysts [2, 11, 32]. The analysis of microphenocrysts certify the datum because it is not
   affected by glass alteration phenomena: the artifacts are normally subjected to alteration burial
   processes, due to waters circulating within sediments [38]. These alteration processes can affect
   alkaline elements mobility, above all Na, and the formation of thin carbonate film incrustations
   [31, 38 and the references therein], sometimes even enlarging the uranium-238 fission tracks [32].
   The obsidian surface alteration can in some way interfere with WD-XRF analysis even making it
   impossible to analyse by SEM-EDS the glass.
   The precision and accuracy of analytical data were extensively discussed for the first time in 1999
   [2] and in other successive papers [11, 31, 32].
   All obsidians from the Maddalena di Muccia site were previously characterised using WD-XRF
   analysis, by measuring the net intensities of some trace elements, namely Rb, Sr, Y, Zr, and Nb [31
   (Table 2).
   3. The reviewer 3 wrote: SEM-EDS confirmation logic is not always clearly structured (Lines 227–
   232; 244–248): SEM-EDS is presented as confirmation of WD-XRF assignments, but the decision logic
   behind when confirmation is required is unclear. The methodological flow between WD-XRF
   screening and SEM-EDS confirmation needs clarification.
   Recommendation: Include a short workflow description explaining analytical decision steps.
   Our replay is: At the beginning of the “Results” section we have changed the sentence “All
   obsidian samples were characterised using non-destructive WD-XRF analysis, which allows
   information on their source area to be obtained very quickly, in less than 30 minutes per sample
   [31], by measuring the net intensities of some trace elements, namely Rb, Sr, Y, Zr, and Nb (Table
   2).” with the sentence:
   “Before discussing the results, it is worth briefly describing the procedure used at the University of
   Bari Aldo Moro, which allows very accurate analytical data, to be obtained with a completely nondestructive
   techniques, useful for attributing the obsidian source areas.
   An initial characterization is performed using WD-XRF, which allows the sample to be
   characterized very quickly, in 30 minutes [31], without having to treat the sample in any way other
   than washing it in a beaker containing distilled water. In this way, the glass and microphenocrysts
   are analysed, and the results are comparable only with laboratories that have used the same
   technique to measure the X-ray intensities of trace elements such as Rb, Sr, Y, Zr, and Nb.
   If the results obtained leave room for doubt, further analyses are carried out using SEM-EDS,
   which allows both the analysis of the major elements of the glass, if it is unaltered, and that of the
   microphenocrysts [2, 11, 32]. The analysis of microphenocrysts certify the datum because it is not
   affected by glass alteration phenomena: the artifacts are normally subjected to alteration burial
   processes, due to waters circulating within sediments [38]. These alteration processes can affect
   alkaline elements mobility, above all Na, and the formation of thin carbonate film incrustations
   [31, 38 and the references therein], sometimes even enlarging the uranium-238 fission tracks [32].
   The obsidian surface alteration can in some way interfere with WD-XRF analysis even making it
   impossible to analyse by SEM-EDS the glass.
   The precision and accuracy of analytical data were extensively discussed for the first time in 1999
   [2] and in other successive papers [11, 31, 32].
   All obsidians from the Maddalena di Muccia site were previously characterised using WD-XRF
   analysis, by measuring the net intensities of some trace elements, namely Rb, Sr, Y, Zr, and Nb [31
   (Table 2).
   4. The reviewer 3 wrote: Archaeological implications are underdeveloped (Lines 298–301; 370–373):
   The manuscript mentions exchange networks but stops short of discussing; Mechanisms of obsidian
   circulation, whether materials arrived as finished tools or raw cores, Potential roles of Muccia within
   exchange networks.
   Recommendation: Expand discussion on settlement function and interaction systems.
   Our replay is: We have added a sentence to the manuscript to clarify this point.
   5. The reviewer 3 wrote: Overly extensive regional literature summary weakens manuscript focus
   (Lines 270–359): The Discussion section contains long summaries of previous data without always
   showing how the new dataset modifies existing interpretations.
   Recommendation: Shorten descriptive regional summaries. Focus on what is newly demonstrated.
   Our replay is: We thank the reviewer for this suggestion. We have added new considerations
   highlighting the significance of our dataset for the reconstruction of obsidian circulation. At the
   same time, we retained a concise summary of previously published regional data, as presenting
   these collectively allows us to document the spatial extent and diachronic development of
   obsidian circulation and to provide a useful framework for interpreting patterns of exchange over
   time.
   6. The reviewer 3 wrote: Alteration and diagenetic processes insufficiently integrated (Lines 210–
   212; 227–229): Alteration is acknowledged but not integrated into analytical interpretation.
   Recommendation: Include short discussion on burial and alteration processes affecting obsidian
   chemistry.
   Our replay is: At the beginning of the “Results” section we have changed the sentence “All
   obsidian samples were characterised using non-destructive WD-XRF analysis, which allows
   information on their source area to be obtained very quickly, in less than 30 minutes per sample
   [31], by measuring the net intensities of some trace elements, namely Rb, Sr, Y, Zr, and Nb (Table
   2).” with the sentence:
   “Before discussing the results, it is worth briefly describing the procedure used at the University of
   Bari Aldo Moro, which allows very accurate analytical data, to be obtained with a completely nondestructive
   techniques, useful for attributing the obsidian source areas.
   An initial characterization is performed using WD-XRF, which allows the sample to be
   characterized very quickly, in 30 minutes [31], without having to treat the sample in any way other
   than washing it in a beaker containing distilled water. In this way, the glass and microphenocrysts
   are analysed, and the results are comparable only with laboratories that have used the same
   technique to measure the X-ray intensities of trace elements such as Rb, Sr, Y, Zr, and Nb.
   If the results obtained leave room for doubt, further analyses are carried out using SEM-EDS,
   which allows both the analysis of the major elements of the glass, if it is unaltered, and that of the
   microphenocrysts [2, 11, 32]. The analysis of microphenocrysts certify the datum because it is not
   affected by glass alteration phenomena: the artifacts are normally subjected to alteration burial
   processes, due to waters circulating within sediments [38]. These alteration processes can affect
   alkaline elements mobility, above all Na, and the formation of thin carbonate film incrustations
   [31, 38 and the references therein], sometimes even enlarging the uranium-238 fission tracks [32].
   The obsidian surface alteration can in some way interfere with WD-XRF analysis even making it
   impossible to analyse by SEM-EDS the glass.
   The precision and accuracy of analytical data were extensively discussed for the first time in 1999
   [2] and in other successive papers [11, 31, 32].
   All obsidians from the Maddalena di Muccia site were previously characterised using WD-XRF
   analysis, by measuring the net intensities of some trace elements, namely Rb, Sr, Y, Zr, and Nb [31
   (Table 2).
   7. The reviewer 3 wrote: Lack of quantitative chronological comparison (Lines 298–301): Statements
   about obsidian distribution changes across periods are not supported by numerical comparisons.
   Recommendation: Add simple quantitative comparisons or charts.
   Our replay is: In consideration of the comment, we have added the sentence clarifying that the
   data in Table 7 derive from studies conducted over an extended period using different analytical
   frameworks, so the proportion of characterised artefacts relative to total assemblages cannot
   always be determined, and the datasets are not statistically comparable. For this reason, summary
   statistics or diagrams by phase were not included.
   8. The reviewer 3 wrote: Figures not always optimally interpreted: Figures 4–9, Several figures are
   only briefly referenced.
   Recommendation: Explicitly interpret figures within the text.
   Our replay is: The figures presented in this manuscript are similar to those presented in many our
   papers published in prestigious international journals, with same captions and references relative
   to our data presented in other previous research [2, 11, 31, 32, 37, 38, 39, 55, 66].
   9. The reviewer 3 wrote: Analytical reproducibility insufficiently described (Lines 142–170):
   Measurement precision and repeatability are not fully documented.
   Recommendation: Include measurement precision or repeatability information.
   Our replay is: At the beginning of the “Results” section we have changed the sentence “All
   obsidian samples were characterised using non-destructive WD-XRF analysis, which allows
   information on their source area to be obtained very quickly, in less than 30 minutes per sample
   [31], by measuring the net intensities of some trace elements, namely Rb, Sr, Y, Zr, and Nb (Table
   2).” with the sentence:
   “Before discussing the results, it is worth briefly describing the procedure used at the University of
   Bari Aldo Moro, which allows very accurate analytical data, to be obtained with a completely nondestructive
   techniques, useful for attributing the obsidian source areas.
   An initial characterization is performed using WD-XRF, which allows the sample to be
   characterized very quickly, in 30 minutes [31], without having to treat the sample in any way other
   than washing it in a beaker containing distilled water. In this way, the glass and microphenocrysts
   are analysed, and the results are comparable only with laboratories that have used the same
   technique to measure the X-ray intensities of trace elements such as Rb, Sr, Y, Zr, and Nb.
   If the results obtained leave room for doubt, further analyses are carried out using SEM-EDS,
   which allows both the analysis of the major elements of the glass, if it is unaltered, and that of the
   microphenocrysts [2, 11, 32]. The analysis of microphenocrysts certify the datum because it is not
   affected by glass alteration phenomena: the artifacts are normally subjected to alteration burial
   processes, due to waters circulating within sediments [38]. These alteration processes can affect
   alkaline elements mobility, above all Na, and the formation of thin carbonate film incrustations
   [31, 38 and the references therein], sometimes even enlarging the uranium-238 fission tracks [32].
   The obsidian surface alteration can in some way interfere with WD-XRF analysis even making it
   impossible to analyse by SEM-EDS the glass.
   The precision and accuracy of analytical data were extensively discussed for the first time in 1999
   [2] and in other successive papers [11, 31, 32].
   All obsidians from the Maddalena di Muccia site were previously characterised using WD-XRF
   analysis, by measuring the net intensities of some trace elements, namely Rb, Sr, Y, Zr, and Nb [31
   (Table 2).
   10. The reviewer 3 wrote: Novelty claim requires stronger framing (Lines 370–373): Claim of first
   identification of multiple Monte Arci sub-sources requires clearer justification.
   Recommendation: Explicitly clarify novelty relative to previous literature.
   Our replay is: We thank the reviewer for this comment. The novelty of our data lies in the absence
   of previous evidence for multiple Monte Arci sub-sources in the Adriatic region, a point we have
   documented and supported with references to previous literature. To make this clearer, we have
   added two sentences in the manuscript explicitly highlighting how our findings expand the current
   understanding of obsidian circulation and source use in the region.
   Minor Comments
   11. The reviewer 3 wrote: English language requires polishing: Several long sentences and minor
   grammatical issues reduce clarity.
   Our replay is: We have revised the English language.
   12. The reviewer 3 wrote: Terminology consistency (Lines 21, 56–60): Use consistent terminology
   for analytical instruments.
   Our replay is: The terminology used for analytical instruments in this manuscript is similar to that
   reported in many our papers published in prestigious international journals [2, 11, 31, 32, 37, 38,
   39, 55, 66].
   13. The reviewer 3 wrote: Table and figure captions could be streamlined (Tables 1, 6): Simplify
   captions.
   Our replay is: The figures presented in this manuscript are similar to those presented in many our
   papers published in prestigious international journals, with same captions and references relative
   to our data presented in other previous research [2, 11, 31, 32, 37, 38, 39, 55, 66].
   14. The reviewer 3 wrote: Introduction slightly repetitive (Lines 48–64): Reduce repetition in
   technique descriptions.
   Our replay is: At lines 53-54 we have cut the sentence: “and Energy Dispersive (ED) Spectrometers
   [14,15].”

Author Response File: Author Response.pdf

Reviewer 5 Report

Comments and Suggestions for Authors

Reviewer Report

Manuscript title:
Obsidian at Maddalena di Muccia (Marche, Central Italy) archaeological site: evidence for three volcanic sources in a multi-phase context

General Evaluation

This manuscript presents a petrochemical and archaeological study of obsidian artifacts from the site of Maddalena di Muccia (Central Italy), based on WD-XRF and SEM-EDS analysis of bulk glass and microphenocrysts.

The paper is scientifically valid, methodologically rigorous, and well contextualised within the regional and Mediterranean framework of prehistoric obsidian trade. The integration of geochemical, mineralogical, and archaeological data provides a robust framework. The identification of multiple Monte Arci sub-sources in the Adriatic area represents an original and significant contribution.

In my opinion, the manuscript is suitable for publication after minor revisions.

Major Strengths

Robust analytical approach. The combined use of WD-XRF, SEM-EDS glass analysis, and mineral chemistry (pyroxene and biotite) provides a highly reliable provenance framework (Sections 2. Materials and Methods, and 3. Results).

Non-destructive methodology. The use of non-destructive techniques is particularly appropriate for archaeological materials and enhances the value of the study.

Clear archaeological contextualisation. The stratigraphic and chronological framework is well described (Section 2.1 The archaeological context), including a balanced discussion of mixing and reworking processes.

Original contribution. The documentation of multiple Monte Arci sub-sources in an Adriatic context offers novel findings with relevant results (Abstract, Section 4. Discussion, and 5. Conclusions).

Extensive regional comparison. The discussion is well supported by a comprehensive review of regional data (Section 4. Discussion).

Major Issues. No major scientific shortcomings were identified. The results are reliable and well supported by the analytical evidence.

Minor Issues and Suggestions. Some aspects would benefit from clarification and minor improvements.

Some typographical and grammatical errors should be corrected, for example:

-Line 21, 143: wavelenght --> wavelength

-Lines 150, 172: becker --> beaker

-Line 167: setted --> set

-Line 178: duplicated with with

-Line 195: Liapri --> Lipari

-Line 272: incorrect bracket […) in citation list

In general, a careful linguistic revision is recommended.

Sampling. (Section 2.1 The archaeological context. Lines 84-85; 104-105; Table 1). The criteria used for selecting the analysed samples are not fully explicit. It would be useful to clarify: how representative the analysed assemblage is; the percentage of the total obsidian collection analysed; whether all major stratigraphic units were systematically covered. A brief clarification would strengthen the methodological transparency.

Chemical Alteration. (Lines 210-212, Table 3). The strong depletion of Na₂O is noted, but its quantitative impact on provenance attribution is only briefly discussed. The authors are encouraged to expand slightly: the effects of burial alteration on glass chemistry; the limits of SEM-EDS glass data under these conditions; how this issue was mitigated by mineral analysis.

Analytical Uncertainty. (Tables 2 - 5; Figures 4, 5, 7, 9). Analytical data are presented without information on uncertainty (standard deviation, instrumental precision, or reproducibility). Even a brief statement on analytical precision would improve the robustness and transparency of the dataset.

4.5 Table 7. Lines 273 - 276

Table 7 is difficult to read due to some formatting inconsistencies. Please, explicit the meaning of “n.d.”, and references should be standardised. A minor revision of this table is recommended.

References. Lines 393 -394. In the reference of Bigazzi, a placeholder reference (“Author 1, A.B.; Author 2, C.”) is present and must be removed before publication.

Conclusions. The conclusions are clear but somewhat conservative. The authors may consider slightly strengthening the emphasis on: the broader implications for Adriatic–Tyrrhenian exchange; the methodological relevance of multi-proxy sourcing; the significance of the Monte Arci sub-source evidence.

Recommendation: Minor Revision

The manuscript is of high scientific quality and makes a valuable contribution to Mediterranean obsidian studies. After addressing the minor issues outlined above, mainly related to language, clarification of sampling, analytical uncertainty, and formatting, it will be suitable for publication.

Comments on the Quality of English Language

Comments on the Quality of English Language

The manuscript is generally clear and understandable. However, it contains some grammatical inaccuracies, typographical errors, and non-idiomatic expressions that occasionally affect fluency and precision. A revision by a proficient English speaker or a professional language editor is recommended to improve clarity, consistency, and overall readability.

Author Response

1. The reviewer 5 wrote: Line 21, 143: wavelenght --> wavelength

Our replay is: We have corrected

 

2. The reviewer 5 wrote: Lines 150, 172: becker --> beaker

Our replay is: We have corrected

 

3. The reviewer 5 wrote: Line 167: setted --> set

Our replay is: We have corrected

 

4. The reviewer 5 wrote: Line 178: duplicated with with

Our replay is: We have corrected

 

5. The reviewer 5 wrote: Line 195: Liapri --> Lipari

Our replay is: We have corrected

 

6. The reviewer 5 wrote: Line 272: incorrect bracket […) in citation list

Our replay is: We have corrected

 

7. The reviewer 5 wrote: Sampling. (Section 2.1 The archaeological context. Lines 84-85; 104-105; Table 1). The criteria used for selecting the analysed samples are not fully explicit. It would be useful to clarify: how representative the analysed assemblage is; the percentage of the total obsidian collection analysed; whether all major stratigraphic units were systematically covered. A brief clarification would strengthen the methodological transparency.

Our replay is: We have clarified that we analysed all five obsidian artefacts recovered from the Early Neolithic feature SU 114, as well as all eight obsidian artefacts from the Copper Age structures. Unfortunately, the single obsidian artefact from the Recent Neolithic pit SU 12 was too altered to allow geochemical characterization analysis.

 

8. The reviewer 5 wrote: Chemical Alteration. (Lines 210-212, Table 3). The strong depletion of Na₂O is noted, but its quantitative impact on provenance attribution is only briefly discussed. The authors are encouraged to expand slightly: the effects of burial alteration on glass chemistry; the limits of SEM-EDS glass data under these conditions; how this issue was mitigated by mineral analysis.

Our replay is: At the beginning of the “Results” section we have changed the sentence “All obsidian samples were characterised using non-destructive WD-XRF analysis, which allows information on their source area to be obtained very quickly, in less than 30 minutes per sample [31], by measuring the net intensities of some trace elements, namely Rb, Sr, Y, Zr, and Nb (Table 2).” with the sentence:

“Before discussing the results, it is worth briefly describing the procedure used at the University of Bari Aldo Moro, which allows very accurate analytical data, to be obtained with a completely non-destructive techniques, useful for attributing the obsidian source areas.

An initial characterization is performed using WD-XRF, which allows the sample to be characterized very quickly, in 30 minutes [31], without having to treat the sample in any way other than washing it in a beaker containing distilled water. In this way, the glass and microphenocrysts are analysed, and the results are comparable only with laboratories that have used the same technique to measure the X-ray intensities of trace elements such as Rb, Sr, Y, Zr, and Nb.

If the results obtained leave room for doubt, further analyses are carried out using SEM-EDS, which allows both the analysis of the major elements of the glass, if it is unaltered, and that of the microphenocrysts [2, 11, 32]. The analysis of microphenocrysts certify the datum because it is not affected by glass alteration phenomena: the artifacts are normally subjected to alteration burial processes, due to waters circulating within sediments [38]. These alteration processes can affect alkaline elements mobility, above all Na, and the formation of thin carbonate film incrustations [31, 38 and the references therein], sometimes even enlarging the uranium-238 fission tracks [32]. The obsidian surface alteration can in some way interfere with WD-XRF analysis even making it impossible to analyse by SEM-EDS the glass.

The precision and accuracy of analytical data were extensively discussed for the first time in 1999 [2] and in other successive papers [11, 31, 32].

All obsidians from the Maddalena di Muccia site were previously characterised using WD-XRF analysis, by measuring the net intensities of some trace elements, namely Rb, Sr, Y, Zr, and Nb [31 (Table 2).

 

9. The reviewer 5 wrote: Analytical Uncertainty. (Tables 2 - 5; Figures 4, 5, 7, 9). Analytical data are presented without information on uncertainty (standard deviation, instrumental precision, or reproducibility). Even a brief statement on analytical precision would improve the robustness and transparency of the dataset.

Our replay is: At the beginning of the “Results” section we have changed the sentence “All obsidian samples were characterised using non-destructive WD-XRF analysis, which allows information on their source area to be obtained very quickly, in less than 30 minutes per sample [31], by measuring the net intensities of some trace elements, namely Rb, Sr, Y, Zr, and Nb (Table 2).” with the sentence:

“Before discussing the results, it is worth briefly describing the procedure used at the University of Bari Aldo Moro, which allows very accurate analytical data, to be obtained with a completely non-destructive techniques, useful for attributing the obsidian source areas.

An initial characterization is performed using WD-XRF, which allows the sample to be characterized very quickly, in 30 minutes [31], without having to treat the sample in any way other than washing it in a beaker containing distilled water. In this way, the glass and microphenocrysts are analysed, and the results are comparable only with laboratories that have used the same technique to measure the X-ray intensities of trace elements such as Rb, Sr, Y, Zr, and Nb.

If the results obtained leave room for doubt, further analyses are carried out using SEM-EDS, which allows both the analysis of the major elements of the glass, if it is unaltered, and that of the microphenocrysts [2, 11, 32]. The analysis of microphenocrysts certify the datum because it is not affected by glass alteration phenomena: the artifacts are normally subjected to alteration burial processes, due to waters circulating within sediments [38]. These alteration processes can affect alkaline elements mobility, above all Na, and the formation of thin carbonate film incrustations [31, 38 and the references therein], sometimes even enlarging the uranium-238 fission tracks [32]. The obsidian surface alteration can in some way interfere with WD-XRF analysis even making it impossible to analyse by SEM-EDS the glass.

The precision and accuracy of analytical data were extensively discussed for the first time in 1999 [2] and in other successive papers [11, 31, 32].

All obsidians from the Maddalena di Muccia site were previously characterised using WD-XRF analysis, by measuring the net intensities of some trace elements, namely Rb, Sr, Y, Zr, and Nb [31 (Table 2).

 

10. The reviewer 5 wrote: Table 7 is difficult to read due to some formatting inconsistencies. Please, explicit the meaning of “n.d.”, and references should be standardised. A minor revision of this table is recommended.

Our replay is: We have inserted in the caption: “n.d. = not determined”

 

11. The reviewer 5 wrote: References. Lines 393 -394. In the reference of Bigazzi, a placeholder reference (“Author 1, A.B.; Author 2, C.”) is present and must be removed before publication.

Our replay is: We have corrected

 

12. The reviewer 5 wrote: Conclusions. The conclusions are clear but somewhat conservative. The authors may consider slightly strengthening the emphasis on: the broader implications for Adriatic–Tyrrhenian exchange; the methodological relevance of multi-proxy sourcing; the significance of the Monte Arci sub-source evidence.

Our replay is: The new Conclusions are: “Petroarchaeometric characterisation of obsidian artefacts of the Maddalena di Muccia site, performed using non-destructive techniques on both the glass matrix and the microphenocrysts, allowed the identification of three different volcanic source areas in the Mediterranean region (i.e. Palmarola, Lipari, and Monte Arci) and, for one of these, even two different sub-sources (i.e. Monte Arci S.A. and Monte Arci S.C.). The study provides, in particular, the first evidence in the Adriatic region of the significant and simultaneous presence of obsidian artifacts originating from multiple sub-sources of Monte Arci, suggesting a lasting and long-term exchange of Sardinian obsidian along with that coming from other Mediterranean sources.

The assemblage of Maddalena di Muccia provides significant data for several reasons. First, they confirm the predominant use of Palmarola obsidian in the Early Neolithic. Second, they show the continued use of obsidian into the Copper Age. Although some artefacts may represent the reuse of materials from older structures, as suggested by double patina, others are clearly associated with the Copper Age settlement phase. Consistent with observations in the Roman area, the quantity of obsidian increases in the final Eneolithic phases.”

 

Author Response File: Author Response.pdf

Reviewer 6 Report

Comments and Suggestions for Authors

Dear Authors, the work presented in draft format, entitled Obsidian at Maddalena di Muccia (Marche, Central Italy) archaeological site: evidence for three volcanic sources in a multiphase context, is exceptionally well designed both formally and conceptually. First, the topic is highly timely, even if it does not represent a novelty with respect to the dating of archaeological artefacts using non-destructive geochemical characterization techniques. Nevertheless, this type of study is truly interesting, as it seeks—through scientific techniques and methods—to provide archaeometric data and, ultimately, to contribute valuable information to the history of civilizations.

With regard to the introductory aspects, the draft is highly appropriate, being both concise and in-depth, and well supported by the relevant bibliography. This is complemented by a thorough explanation of the archaeological context, which—while focusing on the archaeological site and its artefacts from Maddalena di Muccia (northeastern Italy)—also introduces sites geographically distant from the main one, such as the islands of Lipari and Palmarola, as well as the Monte Arci source area in Sardinia. The entire development and its contextualization are presented from an academic and scientific perspective, encompassing both the geolocation of the main site and its artefacts and the experimental design, which includes fundamental techniques for this type of archaeometric characterization. Of particular note is the use of WD-XRF and SEM-EDS techniques in the extensive geochemical quantification, based on trace elements, of the obsidian artefacts that are intended to be culturally interconnected.

In the materials section, the manuscript could be improved if the authors were able to incorporate a graphical presentation consisting of a photographic catalogue of all the specimens studied, including a scale bar to allow for an appropriate visualization of the size of the artefacts and/or their fragments. The authors are kindly requested, insofar as possible, to include such a photographic section documenting the analysed specimens.

The results are presented with considerable thoroughness, both in terms of the contribution of geochemical data per se and the extensive graphical correlations of trace elements (e.g., Y–Nb versus Zr–Nb). In addition, SEM-EDS data are provided concerning the degree of vitrification versus the presence of newly formed pyroxenes in the obsidian artefacts, together with observations on alteration processes and the degradation of mineral inclusions.

Finally, the conclusions section is precise; however, it is requested that the authors consider proposing a hierarchical ordering of the most relevant aspects, ranked from greatest to least importance. The bibliographic references are highly appropriate and specific, as well as abundant.

Kind regards.

Author Response

1. 1. The reviewer 6 wrote: In the materials section, the manuscript could be improved if the authors were able to incorporate a graphical presentation consisting of a photographic catalogue of all the specimens studied, including a scale bar to allow for an appropriate visualization of the size of the artefacts and/or their fragments. The authors are kindly requested, insofar as possible, to include such a photographic section documenting the analysed specimens.

   Our replay is: We have included photos of the analysed obsidian samples, together with a scale bar.

    

   2. The reviewer 6 wrote: The results are presented with considerable thoroughness, both in terms of the contribution of geochemical data per se and the extensive graphical correlations of trace elements (e.g., Y–Nb versus Zr–Nb). In addition, SEM-EDS data are provided concerning the degree of vitrification versus the presence of newly formed pyroxenes in the obsidian artefacts, together with observations on alteration processes and the degradation of mineral inclusions.

   Our replay is: We thank the reviewer for appreciating this aspect of our manuscript.

    

   3. The reviewer 6 wrote: Finally, the conclusions section is precise; however, it is requested that the authors consider proposing a hierarchical ordering of the most relevant aspects, ranked from greatest to least importance. The bibliographic references are highly appropriate and specific, as well as abundant.

   Our replay is: We thank the reviewer for appreciating this aspect of our manuscript. We have reformulated the conclusion in accordance with the reviewer’s suggestion.

    

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report

Comments and Suggestions for Authors

No further comments

Author Response

heritage-4137492-review 3

 

Our replay is: We thank the reviewer for appreciating the revised manuscript.

Author Response File: Author Response.pdf

Reviewer 4 Report

Comments and Suggestions for Authors

The revised version of the manuscript demonstrates a clear and constructive effort to address the comments raised during the previous review and now presents a scientifically sound and well-structured petroarchaeometric study with improved methodological transparency, archaeological contextualization, and linguistic clarity. The WD-XRF and SEM-EDS analytical workflow is more clearly explained, and the discussion of alteration processes and microphenocryst analyses strengthens the reliability of provenance attributions. The manuscript is close to publication readiness; however, its analytical impact could be further enhanced by adding a brief quantitative synthesis of source distributions by chronological phase, such as simple percentage summaries or a concise table, which would allow readers to more readily grasp temporal patterns. In addition, a slightly more explicit interpretative commentary for some figures, a modest tightening of a few regional literature summaries, and a sharper comparative framing of the novelty claim concerning the simultaneous identification of multiple Monte Arci sub-sources would further emphasize the contribution of the study. The English language quality is generally good and requires only minor stylistic polishing in a few longer sentences.

Comments for author File: Comments.pdf

Author Response

heritage-4137492-review report 4, Round 2

 

1. The reviewer 4 wrote: 1. Methodological Transparency and Structural Improvements

Status: Addressed (Satisfactory)

The analytical workflow between WD-XRF screening and SEM-EDS confirmation is now

clearly described, including the rationale for when additional SEM analyses are required. The

integration of alteration effects, Na depletion issues, and the use of microphenocryst analysis

significantly strengthens the methodological robustness.

The archaeological context section has been notably expanded with radiocarbon dates,

stratigraphic clarifications, and a more coherent description of settlement phases and spatial

distribution. These additions effectively respond to earlier concerns regarding contextual

underdevelopment.

Language quality and terminology consistency have also improved, resulting in clearer

readability and a more professional scientific tone.

Our replay is: We thank the reviewer for appreciating this part of the revised manuscript.

 

2. The reviewer 4 wrote: 2. Analytical Synthesis and Quantitative Clarity

Status: Partially Addressed (Further Clarification Recommended)

While the presentation of WD-XRF and SEM-EDS datasets is clearer and figures are now more

explicitly referenced in the text, the manuscript still relies predominantly on descriptive

exposition. Chronological and proportional source distributions are discussed narratively but

not supported by concise numerical summaries or simple statistical charts.

The inclusion of brief phase-based source ratios, summary tables, or proportional diagrams

would significantly enhance analytical clarity without requiring additional data collection. This refinement would allow readers to more immediately grasp temporal patterns and strengthen the evidential weight of provenance interpretations.

Our replay is: As far as this request is concerned: “Chronological and proportional source distributions are discussed narratively but not supported by concise numerical summaries or simple statistical charts”, we have already clarified in the Discussion that: “The data presented in Table 7 derive from studies carried out over an extended period of time and within different analytical and research frameworks; consequently, the proportion of characterised artefacts relative to the total assemblages cannot always be established, and the resulting datasets should not be regarded as statistically comparable.”

Nevertheless, to facilitate the visualisation of temporal patterns, we have added an additional table (Table 8) summarising the absolute number of characterised artefacts by source and chronological phase for sites in the Central Adriatic region of Italy. This table is intended solely as a descriptive overview and does not imply statistical comparability or representativeness of the original assemblages.

 

3. The reviewer 4 wrote: 3. Discussion Depth and Interpretative Framing

Status: Largely Addressed (Minor Refinement Suggested)

The Discussion section has been shortened and is more focused compared to the previous

version, and archaeological implications, particularly exchange networks and site roles, are now

more explicitly articulated. The integration of alteration processes and diagenetic

considerations into the interpretative narrative represents a clear improvement.

However, certain regional literature summaries remain slightly extensive, and the novelty claim

regarding the simultaneous presence of multiple Monte Arci sub-sources could benefit from

sharper comparative framing relative to earlier studies. A modest tightening of literature

synthesis and clearer emphasis on how the new dataset modifies existing regional models would

further enhance the manuscript’s contribution.

Our replay is: We thank the Reviewer for this comment. We consider it important to include the relevant regional literature to provide appropriate context for the study. The manuscript has been revised to focus the discussion on the most relevant studies and to highlight more clearly the significance of the new results. In our view, these revisions sufficiently address the point, and we do not consider further expansion of the discussion necessary.

Author Response File: Author Response.pdf