The present paper by Richirt et al. reports on the discovery of an opal layer along the inside of the calcitic tests of cosmopolitan benthic species Bolivina spissa. The authors used a variety of electron microscopy techniques to document the opal layer as well as the putative organelles involved in its precipitation. The authors conclude that the opal layer was precipitated by the foraminifera itself, in a manner analogous to the precipitation of the calcitic test, and that this layer is most likely a defence mechanism against predation. Additionally, the authors suggest that the presence of an opal layer might 1) offer a novel paleoclimate proxy based on its stable isotope composition and 2) negatively affect paleoclimatic reconstructions based on the bulk composition of the test.

Overall, the manuscript provides a robust investigation into the morphology, composition and function of a bi-mineralic foraminifera test. The introduction reads very well and serves as an introduction to biosilification generally, the importance of foraminifera in paleoclimatic reconstructions, and as a good historical overview of silification in foraminifera. The figures (and supplementary figures) presented in the manuscript are generally very well organised and present a very convincing case for the presence of the opal layer. Section 4.2 is particularly well-written, with the numbered list concisely laying out all the arguments for why the authors think the opal layer is secreted by the foraminifera itself and not due to environmental or passive processes. The TEM images and documentation of the putative SDVs found in B. spissa are an exciting finding and it was very smart on the part of the authors to compare images of the putative SDVs in B. spissa with SDVs in another silicifying organism. The argumentation in the discussion follows a logical structure, and the conclusions reached on the presence of the opal layer, its presumed precipitation mechanism and its function are very compelling. There are some minor problems with the use of English, but these did not distract from the pleasure of reading a well-structured and clearly written manuscript. Suggestions for improving the language in the manuscript are provided in the line-by-line comments attached.

The only major issue with the paper concerns the isotopic data and section 4.4: Implication(s) for palaeoproxies and biogeochemical cycles. Firstly, how were the samples cleaned to remove potential contamination from sedimentary particles and organic contaminants particularly, which can interfere with the CO₂ produced during analysis of stable isotopes during bulk analysis? Standard cleaning typically involves ultrasonication in methanol, deionized water and boiling with hydrogen peroxide to remove organic contaminants. See Roberts et al. (2018) for an overview on cleaning methods.

Secondly, the authors report lower d¹⁸O and d¹³C values on the proloculus side compared to the aperture side and relate this to the thickness of the opal layer. However, looking closely at the d¹⁸O data for example, only seven samples have data for both sides. 3 tests show the proloculus side lower than the aperture side by a difference significantly greater than the analytical precision of the mass spec (±0.1 ‰), 1 test shows a small difference, 2 tests show no significant difference and 1 test shows the reverse trend. If the authors wanted to seriously look at differences in the stable isotopic composition from the proloculus to the aperture side and use it as evidence for the effect of the opal layer on paleoclimate reconstructions based on stable isotopes, SIMS or LA-ICP-MS measurements of each chamber would be more appropriate. Additionally, as stated in the manuscript, these differences could be related to ontogenetic differences or calcification at different seawater temperatures, so a study looking into this would need to be carefully designed to account for these factors.

Third, while discussing how the presence of the opal layer might potentially influence proxy calibrations and their interpretation by mixing of the calcite and opal signals, the authors cite Borelli et al. (2018) who suggested that the presence of incorporated silicate grains within the tests of another benthic foraminifera species might affect paleoclimate reconstructions. However, both Borelli et al. and the present manuscript fail to account for the fact that the internal Si-O bond requires significantly more energy to break, and therefore extremely powerful oxidising reagents (fluorine-based compound such as ClF₃ or BrF₅) are required for the analysis of oxygen isotopes in silica-based skeletons such as diatom frustules (Leng and Barker, 2006). Therefore, it appears unlikely that the presence of an opal layer would influence paleoclimate reconstructions based on the bulk oxygen isotope compositions of foraminifera tests that are typically extracted using phosphoric acid. Removal of this section or alternatively, a more robust evaluation of this claim would improve the manuscript.

Finally, considering how thin the opal layer in B. spissa is shown to be in this study, and particularly coupled with the fact that, as the authors themselves state, this opaline layer has not been found in other very well-studied foraminifera, the evidence is thin for suggesting that Foraminifera could play a 'substantial' role in silicon cycling. The language in this section could therefore be softened somewhat. Suggesting the use of the opal layer as a new paleoproxy is already a significant and novel contribution that potentially opens up a new field of study and is a good place where to end off the manuscript.

L.R. Roberts, J.A. Holmes, M.J. Leng, H.J. Sloane, D.J. Horne, Effects of cleaning methods upon preservation of stable isotopes and trace elements in shells of Cyprideis torosa (Crustacea, Ostracoda): Implications for palaeoenvironmental reconstruction, Quaternary Science Reviews, Volume 189, 2018, Pages 197-209, https://doi.org/10.1016/j.quascirev.2018.03.030.

Melanie J. Leng, Philip A. Barker, A review of the oxygen isotope composition of lacustrine diatom silica for palaeoclimate reconstruction, Earth-Science Reviews, Volume 75, Issues 1–4, 2006, Pages 5-27, https://doi.org/10.1016/j.earscirev.2005.10.001.

Dear editor,

I carefully read the manuscript you provided me with from Richirt and co-workers on layers of opal found coating the inside of foraminiferal shells. In short, I think this is a fine piece of work, of broad interest and can be published with minor corrections. Below, I outlined a few issues that can be clarified to turn this manuscript into a very nice publication!

Sincerely,



Lennart de Nooijer

Minor issues:

Line 36: ‘sponges or protists’ can be ‘sponges and protists’

Line 56: ‘from the North Pacific’

Line 57: ‘investment’ is a bit odd here: consider replacing by ‘coating’ or something similar.

Line 78: ‘has’

Line 99: ‘resumes’ can be ‘summarizes’

Line 149: ‘aspect’? Replace by ‘appearance’

Line 160: ‘prior to the measurements’

Line 180: ‘connections’

Line 247: I don’t understand the ‘aspect’ here.

Line 289: This is not necessarily the case. It could easily well be that the formation of this inner opal layer is continuous and that the decrease in thickness with size (figure 6) reflects simply the changes in surface-to-volume ratio. Now I think about it: if they would have added a little layer with a constant rate, the decrease in thickness with size would probably describe a different curve (e.g. smaller chambers would have a much more similar thickness than larger chambers). It may be, as the authors suggest, that a little layer is added with each chamber addition event, but I don’t see how this would lead to the power function of figure 6. Anyway, the ‘ontogenetic effect’ (line 289) is a bit out of place here. There is a trend with size, and according to the fitted curve, some explanations are more likely than others. I think this paragraph could be extended: what is the size/weight or surface/volume curve for Bulimina’s of different ages look like and could they indicate what controls the opal coating and the change in its thickness?

Line 307-308: this is a very interesting observation! I think the Methods and Results sections should explicitly mention that also other species were analyzed, but did not show the opal coating. This is important for the discussion.

Line 364-366: To me, the co-occurrence of opal and calcite in these specimens means that the formation of opal is not (necessarily) an ‘alternative’ to calcification.

Lines 410-413: could it also be that the formation of this opal layer is an ‘unwanted’ by-product of another process? It has been suggested that calcification as such (in foraminifera) started as a way to get rid of intracellular, harmful [Ca²⁺].

Lines 421-423: I don’t understand this: if most of the opal is secreted áfter the calcite is formed, how can that process affect the del-18O of the calcite? Which is fixed at the time of calcification.