The primary standards and peer-reviewed work behind the Grit Table
Every claim in the table traces to a source here. Take them, cite them, check our work.
Primary and peer-reviewed literature behind the Grit Table claims. Each entry gives a full citation, an accessible URL, paywall status, and one line on which Grit Table claim it supports. The dataset so far was sourced from retailers and a popular book (Schwarz, Sharpen This); this file grounds the core claims in standards and peer-reviewed work.
Conventions: "Paywalled" means the full text costs money (ISO, JIS, ANSI, and most journal articles). "Free" means an open PDF or open-access page exists. Where a standard is paywalled but a vendor or distributor mirrors the numeric tables for free, that free mirror is noted.
1. Particle-size metrology: a single number is not enough, a distribution is
These back the Atlas "spread" model: abrasive grains are irregular and non-spherical, so one size cannot describe a grain. A particle-size distribution (and an explicit choice of equivalent diameter) is required.
Jillavenkatesa, A., Lum, L. S. H., Dapkunas, S. J. (2001). NIST Recommended Practice Guide: Particle Size Characterization. NIST Special Publication 960-1, National Institute of Standards and Technology, Gaithersburg, MD. DOI: https://doi.org/10.6028/NBS.SP.960-1. Free PDF: https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=850451. Free (US government publication). Supports: the spread model. Introduces the equivalent (spherical) diameter precisely because real particles are non-spherical, and states that a measured particle-size distribution depends on the instrument and technique used. Best single open-access anchor for both this section and section 2.
Allen, T. (1997). Particle Size Measurement, 5th ed. Powder Technology Series, Chapman and Hall / Springer. Springer chapter "Particle size, shape and distribution": https://link.springer.com/chapter/10.1007/978-94-009-0417-0_4. Borrowable scan: https://archive.org/details/particlesizemeas0000alle. Partly paywalled (Springer chapter paid; Internet Archive copy borrowable free). Supports: the spread model and the methods catalogue. The canonical reference for equivalent-sphere diameters (sieve, Stokes, projected-area, volume) and the argument that no single number describes an irregular grain.
ISO 9276-6:2008. Representation of results of particle size analysis, Part 6: Descriptive and quantitative representation of particle shape and morphology.https://www.iso.org/standard/39389.html. Paywalled (read-only preview at https://www.iso.org/obp/ui/#iso:std:iso:9276:-6). Supports: the spread model. The ISO standard that formalizes the shape and morphology descriptors needed beyond a single size for irregular particles.
ISO 9276-2:2014. Representation of results of particle size analysis, Part 2: Calculation of average particle sizes/diameters and moments from particle size distributions.https://www.iso.org/standard/57641.html. Paywalled. Supports: the spread model. Defines how mean diameters are computed from a distribution, which makes the distribution (not a single value) the object of description.
2. Measurement methods, and why they disagree
These back the claim that grain size is measured by several distinct methods (test sieving, sedimentation, laser diffraction, electrical sensing zone / Coulter, SEM or optical image analysis), each reporting a different equivalent diameter, so the methods do not agree on one number. Every ISO method standard below defines one of those equivalent diameters.
ISO 13317-1:2024. Determination of particle size distribution by gravitational liquid sedimentation methods, Part 1: General principles, requirements and guidance.https://www.iso.org/standard/78456.html. Paywalled. Supports: sedimentation (Stokes / settling diameter). Microgrit abrasives are graded by sedimentation, so this is method-relevant to the fine end.
ISO 13322-1:2014. Particle size analysis, Image analysis methods, Part 1: Static image analysis methods (microscopy and SEM image analysis). https://www.iso.org/standard/51257.html. Paywalled. Supports: SEM / optical image analysis (projected-area or equivalent-circle diameter).
Hyslip, J. P. comparison work and the soil-science method-comparison literature. Representative open-access papers that quantify how methods disagree on the same sample:
"Determination of Particle Size Distribution: Comparison of Standard Hydrometer Method and Laser Diffraction Analysis." Forests 15(2):327 (2024). https://www.mdpi.com/1999-4907/15/2/327. Free (open access).
"Adequacy of laser diffraction for soil particle size analysis." PLoS ONE. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5417517/. Free (open access). Supports: methods disagree. Peer-reviewed evidence that sedimentation and laser diffraction report different size thresholds for the same sample because non-spherical particles are measured differently. Use these when a peer-reviewed (not standards or vendor) "methods disagree" citation is wanted.
Microtrac, "Different Particle Analysis Techniques Compared."https://www.microtrac.com/applications/knowledge-base/different-particle-analysis-techniques-compared/. Free (vendor knowledge base, not peer-reviewed). Supports: methods disagree. A clean worked demonstration that sieving, laser diffraction, and image analysis give different results for the same irregular powder. Useful for illustration, not as a primary citation.
Note: no single peer-reviewed sentence enumerates exactly these five named methods together. The list is assembled from the five ISO method standards above plus NIST 960-1 (which covers four of the five; the electrical sensing zone is covered by ISO 13319 rather than by NIST 960-1). If a single document is wanted, Allen's textbook covers all five.
3. Abrasive grit grading standards (the grit-to-micron definition)
These are the primary definitions behind the Atlas tier A "standard" rows. Each grit number is a label defined by one of these systems, and the systems disagree at the fine end. ISO 8486 equals FEPA-F (bonded), ISO 6344 equals FEPA-P (coated): ISO publishes the test method, FEPA originated the grades. Cite ISO for the formal definition, FEPA or the free charts for the numeric tables.
ISO 6344-3:2021. Coated abrasives, Part 3: Microgrit sizes P240 to P5000.https://www.iso.org/standard/78219.html. Paywalled. Supports: the fine P-grade microgrit definition (the sandpaper fine end, the one that diverges most from the bonded and JIS systems). Predecessor covering P240 to P2500: ISO 6344-3:2013, https://www.iso.org/standard/56010.html.
JIS R 6001:1998. Bonded abrasive grain sizes (Japanese Industrial Standard; the JIS # system that waterstone grits are quoted in). Japanese Standards Association. Sold in English via ANSI Webstore: https://webstore.ansi.org/standards/jis/jis60011998. Paywalled. In 2017 the standard was split into JIS R 6001-1:2017 (macrogrits) and JIS R 6001-2:2017 (microgrits), harmonized toward ISO 8486: https://webstore.ansi.org/standards/jis/jis60012017. Supports: the JIS # number-to-micron definition. Note that the 2017 harmonization means modern JIS # is close to but not identical with legacy JIS numbers.
ANSI/UAMA B74.18-2018. Grading of Certain Abrasive Grain on Coated Abrasive Material (the CAMI grit system for US sandpaper). Unified Abrasives Manufacturers Association. https://store.accuristech.com/standards/ansi-uama-b74-18-2018. Publisher: https://uama.org/publications/. Paywalled. Supports: the CAMI coated-abrasive definition (the US "400 grit" near 22 microns, far finer than FEPA P400).
Grit Size Conversion Chart, Metallographic and Materials Society (metallography.org).https://www.metallography.org/resources/grit-size-chart. Free. Cites FEPA 43-GB, ANSI B74.18, and JIS R6010 as its basis. Supports: both the standards-define-the-mapping claim and the systems-disagree-at-the-fine-end claim. Independently corroborates the README figures: it lists ANSI/CAMI 400 at 22 microns, FEPA P400 at 35 microns, JIS 400 at 32 microns, and states the systems are not a 1:1 conversion past roughly 240 grit. The README "CAMI 400 is about 23, FEPA P400 is about 35" is within rounding of these published values.
4. Sharpening science (SEM edge studies and honing tribology)
Verhoeven, J. D. (2004). Experiments on Knife Sharpening. Emeritus Professor, Department of Materials Science and Engineering, Iowa State University, Ames, IA. September 2004. PDF: https://northarmknives.com/wp-content/uploads/2016/01/knifeshexps.pdf. Mirror: https://archive.org/details/experiments-on-knife-sharpening (Internet Archive). Free. A self-published technical report, not formally peer-reviewed, but a rigorous SEM study and the most-cited primary source in amateur sharpening metallurgy (Verhoeven co-authored the peer-reviewed Damascus-steel papers). Supports several Atlas claims directly, from his own SEM imaging:
A finer advertised grit does not guarantee a finer abrasive action. His 6000 and 8000 grit Japanese waterstones produced essentially the same edge, and the grooves were "perhaps slightly finer on the supposedly coarser 6000 grit stone." His words: "an advertised finer grit in this 6000 to 8000 size range does not guarantee a finer abrasive action." This is the strongest single primary anchor for the Atlas thesis that fine-grit numbers are unreliable.
The stone or wheel sets the edge and leaves a bur; the final polish removes it. Of the three honing methods tested, the best was a leather strop loaded with chromium oxide or diamond compound, giving edge widths of 0.3 to 0.5 microns, matching razor-blade standards.
Particle sizes are stated for the compounds: 0.5 micron chromium oxide, and 1 and 6 micron diamond aerosol sprays. The 1 micron diamond gave optimum edge widths near 0.3 microns; the 0.5 micron chromium oxide gave near 0.4 microns; little improvement was found going from chromium oxide to 1 micron diamond or from 6 to 1 micron diamond. Supports the Atlas point that beyond a certain fineness, going finer stops mattering for the edge.
A 1200 grit DMT diamond hone (diamond embedded in a nickel matrix on a steel plate) produced "surprisingly large surface abrasion marks," rougher than the waterstones. A relevant nuance for the Atlas diamond category: a plated diamond hone behaves coarser than its grit label suggests, separate from diamond paste.
Burr formation is an inherent feature of stone or wheel sharpening; finer media leave a smaller bur (around 0.5 microns off the waterstones, 1 to 1.5 microns off 1000 grit wheels, 2 to 4 microns off 200 grit wheels).
5. Japanese natural stones (the ~2 to 3 micron radiolarian-silica figure)
Science of Sharp, "JNAT Slurry, Part 2" (2016).https://scienceofsharp.com/2016/03/29/jnat-slurry-part-2/. Free (a blog, but primary SEM/EDX work by the author). SEM and EDX of slurry from four Nakayama stones. Supports: both the particle-size figure and the composition. Direct observation: "silica particles of approximately 3 microns are visible, dimensions and shapes typical of Radiolaria silica." Identifies the silica (radiolarian) particles as the abrasive and the phyllosilicate clay as binder. The most defensible single combined anchor for the Japanese-natural claims, and primary (own imaging).
Science of Sharp, "Does J-nat Slurry Break Down?" (2014).https://scienceofsharp.com/2014/05/28/does-j-nat-slurry-break-down/. Free. Earlier SEM/EDX work identifying the silica plus silicate phases. This is also the page that carries the "Masumura Tajiro 1979, 2 to 3 micron" attribution, in a reader comment (see the caveat below).
Mikami (Mikakami), T., Mukai, K., Tokura, N., Imoto, N. (2002). Relationships among quartz crystallinity index, grain size of quartz, and conodont color alteration index of bedded cherts in the Tamba Belt, Southwest Japan. Journal of the Geological Society of Japan (Chishitsugaku Zasshi), 108(12), 806-812. DOI: https://doi.org/10.5575/geosoc.108.12_806. J-STAGE: https://www.jstage.jst.go.jp/article/geosoc1893/108/12/108_12_806/_article. Free (open access, Japanese with English abstract). Supports: the geology and the grain-size figure. Peer-reviewed measurement of quartz grain size in the Tamba belt bedded cherts, the exact geological formation the Kyoto and Tanba finishing stones come from, with quartz grain size in the low-single-digit micron range (correlated with thermal history). The closest peer-reviewed number to the "2 to 3 micron" claim, arrived at independently.
Matsuda, T., Isozaki, Y. (1991). Well-documented travel history of Mesozoic pelagic chert in Japan. Tectonics, 10(2), 475. https://ui.adsabs.harvard.edu/abs/1991Tecto..10..475M/abstract. Abstract free, full text paywalled. Supports: the radiolarian origin. Establishes that the Mino-Tamba bedded cherts are Triassic to Jurassic pelagic radiolarian sediments, the geological basis for calling the abrasive "radiolarian silica."
Bernal Cutlery, "Natural Stone Primer."https://bernalcutlery.com/pages/natural-stone-primer. Free (retailer, not peer-reviewed). Supports: the radiolarian-silica narrative in plain language. Use only for framing; for age trust the peer-reviewed geology above (Bernal says Cretaceous, the geology dates the chert to Triassic to Jurassic).
Caveats and weak anchors (for softening Atlas copy)
"Masumura Tajiro 1979, 2 to 3 microns" is not a verified primary source. It traces only to a reader comment on the Science of Sharp 2014 post, attributing the figure to a popular booklet ("The Charm of The Stones of Kyoto"), not a journal article. The romanized name does not resolve to any indexed author, and no library record or digitized copy was found. The commenter himself notes uncertainty about where the samples came from. Do not present "Masumura 1979" as a checked primary figure. Cite the number to Science of Sharp 2016 (about 3 microns, own SEM) and to Mikami et al. 2002 (peer-reviewed quartz grain size in the same formation), and mention Masumura only as "popularly attributed."
The exact five-method enumeration (sieving, sedimentation, laser diffraction, electrical sensing zone, image analysis) is assembled from the five ISO method standards rather than quoted from one peer-reviewed sentence. Allen's textbook is the closest single source that treats all five. Safe to state as standards-backed; do not attribute the five-way list to one paper.
The FEPA macrogrit-by-sieving versus microgrit-by-sedimentation split is reported by FEPA summaries and vendor pages, not quoted here from the controlled standard text (the full FEPA documents are purchase only). State it as FEPA-summarized.
Verhoeven (2004) is a self-published report, not formally peer-reviewed. Frame it as a rigorous SEM study by an emeritus materials-science professor, which is how it is treated in the field, rather than as a peer-reviewed journal paper.