Fingerprinting Paranesti Rubies through Oxygen Isotopes

Round  1

Reviewer 1 Report

The study deals with a description of O isotopes in a new locality from Greece. I took this study to mean gemmological significance and fingerpriniting without expecting much of an interpretation part. Interpretation part is allright. English is good.

Line 18 …and represent one of the lowest values yet recorded for ruby.

            This is an overstatement (give Karelia examples that you quote) and can be dropped. I suggest emphasizing hydrothermal origin.

Lines 31-37 too general an intro. May be Minerals readership does not know that.

Rest of Intro. It is not surprising to find high d18O values in rubies as most are of sedimentary origin. Rather than “fingerprinting” emphasize isotope fractionaton as a function of temperature and origins: weathering vs high-T hydrothermal.

Fig 4 surely beautiful crystals!

Lines 90-100 is clear description of methods.

Section 4.3 Again, I would start with 1000lna-T plot for corundum for water-corundum, quartz corundum, pyroxene corundum, indicate the range of temperatures of expected formation of protoliths, and ranges of waters, then I would compute equilibrium Al2O3. You may use increment method of Zheng 1993 as o direct experiment with Al2O3 exists to my knowledge.

Table 4 misses on Bindeman et al 2010 Geology paper. 

Main comment: On Fig 5 please directly show measured d18O! This would show the readers the range of natural variations. Remove hand-written spot numbers.

In interpretations, hydrothermal or different types of water-rock interaction model is the only one plausible. With values of d18O nothing extraordinary is required. Any meteoric water with d18O >-10 ‰ that are abundant on 90% of continental surface (waterisotopes.org) are able to serve as parental fluid that alters protolith:

1000lna(cor-water) = d18Ocor - d18Ow; thus d18Ocor =1000lna(cor-w) + d18Ow. 

The cor (corundum) in this equation can be originally kaolinite-something mix with excess Al.

If space is in issue for this journal, the text can be shortened by reporting observations, SIMS zoning pictures (important!) and say that hydrothermal origin is preferred. 

Author Response

Dear reviewer 1,

Thank you for taking the time to review my manuscript just before the Christmas holidays. Our responses have been delayed due to most of the authors from Australia have been on summer vacations. We are all back now.

Please find attached a document addressing all the review points with responses in red font.

Many thanks

Kandy

Author Response File: Author Response.pdf

Reviewer 2 Report

The manuscript by Wang et al. provides new interesting results of low d18O rubies from Paranesti, Greece. The results are important for isotopic tracing of rubies and will make an important contribution to the global ruby database. The two main concerns I have with the manuscript are: (1) an incomplete treatment of the analytical data and assessment of the accuracy and precision (detailed comments below), and (2) section 4.3 (causes of low d18O values) does not have a strong conclusion and should be reorganized to place the least likely causes early and the most consistent model towards the end. This would then link better with the conclusions. 

Specific comments:

L72: How specifically were the rubies cleaned? With nitric acid? In an ultrasonic bath?

L74: What are the inclusions?

L84-89. The methodology is complete, but there are two important aspects missing: (1) Was a secondary standard run to verify accuracy? If so, what are the accepted and measured values. (2) What is the reproducibility of the data? Without these two pieces of information, we cannot verify the accuracy and precision of your data.

L91-99: Same point as for the laser fluorination. What secondary standards were run on the SIMS and what were the  results. Without this information, we cannot verify the accuracy and precision of your data. Your primary standard (PAR-1) was used to correct for instrumental mass fractionation, but how do we know that the correction you applied is equally valid for all your unknowns? This is why secondary standards are essential. 

L97: Did you examine 3 fragments from the same ruby grain or 3 fragments from different ruby grains?

L121: You cannot say that the SIMS data are highly precise nor accurate without reporting the measured values and reproducibility of secondary standards. Your spread of values on PAR-1 could be from instrumental issues or sample heterogeneity. Therefore, you cannot use this to assess your precision. 

L125: What is meant by "location proximity"?

L146: "centred" is not a proper statistical term. Consider plotting your data as histograms and then you can talk about the distribution of the data and the most common value (i.e. mode). 

L151-152: Interesting, plot this information directly on the photo in Figure 5c. 

L153: What uncertainties? Please provide what the uncertainty is on your analyses in the methods section. 

L180: What is meant by "correspond up to an unknown geological sapphire type"? This wording is confusing. 

L197: "theory" is misleading. "model" is better or even "hypothesis". 

L200-201: I don't follow the argument of this sentence following on from the last. Why does an absence of inclusions and a metamorphic origin preclude it from being isotopically similar to the protolith? Metamorphism does not always affect the d18O of a rock. 

L211: "inconclusive" why? Or is the model inappropriate because the Paranesti rubies are amphibolite and not granulite facies?

L231-232: Your rubies are thought to have been produced at amphibolite-facies conditions (>500C). Do you expect significant kinetic isotope fractionation at these temperatures? I would not think so. 

L245: Did you measure the values of the host rock to see if it has low d18O values? If not, then you cannot preclude this possibility. 

Figure 1 is not very clear. It is hard to read some of the place names. Make sure that the names are legible in the final version. 

Figure 4c requires a scale. 

Figures 5a and 5b are not necessary. 

Figures 5c and 5d: the markings are not legible. Mark the spot locations and also the values so that the reader can evaluate if there is significant core-rim zoning.

Author Response

Dear reviewer 2,

Thank you for taking the time to review my manuscript just before the Christmas holidays. Our responses have been delayed due to most of the authors from Australia have been on summer vacations. We are all back now.

Please find attached a document addressing all the review points with responses in red font.

Many thanks

Kandy

Author Response File: Author Response.pdf

Round  2

Reviewer 1 Report

ok after revisions