Calcite-Mineralized Fossil Wood from Vancouver Island, British Columbia, Canada

Round 1

Reviewer 1 Report

The manuscript is very well written and illustrated and can be published as submitted.

There only one detail that I encourage the authors to change: replace the term "ground water" with "pore fluids".  The term ground water connotes the occurrence of fresh water moving through Earth surface materials. The fossils described were deposited in marine water-saturated sediments, a very different kind of water medium. Pore fluids is a more general term that is appropriate here.

Author Response

Thanks for the encouraging review. I have adopted your advice about "pore fluids" versus "groundwater".

Reviewer 2 Report

Manuscript ID: geosciences-1081690
Type of manuscript: Article
Title: Calcite-mineralized fossil wood from Vancouver Island, British
Columbia, Canada
Authors: George E. Mustoe *, Graham Beard

There are many positive points about this article, and I recommend publication.  The relatively uncommon occurrence of calcite permineralization is very well documented.  Excellent photomicrographs.  That said there some negative aspects. 

It reads like a MS thesis, padded out with some interesting but largely irrelevant figures and text.  With some text adjustments authors could omit Figures 2, 3, and 9.  Figure 7, EDS spectra, needs to be fully quantified with an appropriate standard curve.  (I taught our SEM/EDS course for many years and EDS peak ratios are subject to a great deal of variation depending on the area scanned.)

The section on concretions is marginal to the main theme, calcite permineralization, and seems rather detailed and long.

Minor point.  Line 138 refers to Francis 1988 rather than a citation number (12?)

Author Response

Thanks for the helpful and constructive review. If I were submitting this manuscript to a traditional printed journal were space limitations are an important issue I would have written a more condensed version. One of the benefits of online publication is that page length is less important. I have written in a story-telling style because I believe that it will make the article more accessible to readers who are not mneralogiists, geochemists, or paleobotanists. I have eliminated Figure 9, but left the other figures based on the recommendations of the associate editor.

Your concerns regarding Figure 7, the SEM/EDS spectra, merits some explanation. Like you, I spent many years teaching SEM/EDS, and I also have deep experience with XRF bulk quantitative analysis of geologic materials using both wavelength-dispersive and energy-dispersive spectrometers. As I noted in the text, it is not possible to obtain quantitative data for specimens that contain carbonate minerals or carbonaceous organic matter. The reasons are that the 0.277 C peak is difficult to detect because of its very low energy, and because the XRF emission ("fluorescence yield") is very low. As a result, the height of the carbon peak gives an approximate indication of elemental abundance, but attempts at quantification are never successful. I base that statement on my experience with both traditional mylar-windowed lithium-drifted Si  detectors, and a new generation of SDD detectors. In both instances, calculated values for carbon are at least double the actual value.

You advocate listing standard curves, but this is not a feasible option. Although standard curves work pretty well for the high resolution peaks obtained for major element with the wavelength-dispersive XRF spectrometers that are used for bulk analysis and on some electron microprobes, the energy-dispersive detectors used in SEM's require a very different approach. The consideration of absorption/enhancement effects requires multivariate mathematics that cannot be represented as simpler standard curves. Also, SEM/EDS systems typically use a standardless approach where quantification is based on X-ray theory, based factors that include the atomic number and known absorption/fluoresecence properties of each element, as well as the beam voltage and takeoff angle. In the EDAX and Oxford systems  that I have used, the final quantification is based on the normalization of the results to total 100%. This means that if an element like carbon is mismeasured, the error is distributed over the results for every other element. It may be a software option to calibrate the calculations to fit data from standards of known composition, and approach we have sometimes used for mineral analysis, but it does not solve the inability of detectors to accurately quantify carbon.

In regards to your concern about the variability of results when individual spots are analyzed on an SEM specimen, I agree that this is a subject with consideration. However, in this case the calcite-mineralized woods have compositions that are fairly homogeneous. It's pretty different from many geologic samples, where micromineral assemblages can produce localized compositional differences. The spectra I used in Fig. 7 are representative for the various specimens, all of which were subject to analysis of multiple spots.

Thanks again for the careful review.