Pith. sign in

REVIEW 4 major objections 4 minor 37 references

Galaxy Morphology Classification: Are Stellar Circularities Enough?

T0 review · 4 major / 4 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read Stellar orbital circularity alone can split simulated galaxies into early-, late-type, and irregular systems, and a disk-fraction threshold of 0.25 recovers the observed morphology-density relation.

desk verdict A useful, honestly-labeled preliminary study showing circularity-based disk fractions track thin disks in TNG50 and proposing a lower, data-motivated threshold for TNG100, but the threshold and the morphology-density claim rest on the same distribution and need sensitivity tests. read the letter →

arxiv 2506.20002 v1 pith:RTDTJL2U submitted 2025-06-24 astro-ph.GA

Katarina Baucalo , Ana Mitrašinović This is my paper · ORCID
classification astro-ph.GA
keywords galaxymorphologystellarcircularitykinematicdecompositionIllustrisTNGmorphology-densityrelationearly-typegalaxieslate-typediskfraction
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper argues that a single per-star kinematic number — the orbital circularity already tabulated in a public catalog that accompanies the IllustrisTNG simulation suite — is enough to classify simulated galaxies into early-type, late-type, and irregular or morphologically complex systems, without running expensive five-component orbit decompositions or producing mock images. The claim is grounded in a validation against a detailed five-component decomposition in TNG50, which shows the circularity-based disk fraction tracks the thin disk and the spheroid fraction tracks classical bulge plus stellar halo. From a two-peaked distribution of disk fractions in intermediate-mass galaxies within TNG100 at $z=0$, the paper sets the early/late boundary at $F_{\rm disk}=0.25$, and marks systems with disk plus spheroid fraction below $2/3$ as irregular. Applying this scheme recovers a broad morphology-density relation: late-type galaxies dominate the field, early types dominate clusters, in line with observations. The payoff, if the claim is right, is a cheap and scalable morphology label covering every snapshot and simulation volume, useful for statistical studies and for machine-learning training sets.

What carries the argument

The load-bearing object is the stellar circularity parameter $\varepsilon = j_z/j_c(E)$, defined as a star's angular momentum component along the galaxy's rotation axis divided by the angular momentum of a circular orbit at the same binding energy; it runs from $-1$ for counter-rotating orbits to $1$ for aligned circular orbits. From this single quantity the public catalog yields two scalars: the disk fraction $F_{\rm disk}$ (the mass fraction of stars with $\varepsilon > 0.7$ minus the fraction with $\varepsilon < -0.7$) and the spheroid fraction $F_{\rm sph}$ (twice the mass fraction below $\varepsilon=0$). These are validated against a five-component kinematic decomposition in TNG50, then used to build the classification: the two-peaked $F_{\rm disk}$ distribution over $10 \le \log(M_\star/M_\odot) \le 12$ fixes the early/late split at 0.25, the condition $F_{\rm disk}+F_{\rm sph}<2/3$ flags irregular or complex systems, and environment is a three-tier binning into field, group, and cluster based on host halo mass and subhalo count. The machinery turns millions of stellar orbits into two accessible numbers that can classify galaxies and connect to environmental trends.

What would settle it

Recompute the disk-fraction distribution in TNG100 at $z=0$ in narrow stellar-mass bins: if the two-peaked shape with a minimum near 0.225 appears only in the $10$–$12$ log-mass range, or if the minimum shifts outside roughly 0.15–0.35 when resolution, redshift, or mass selection changes, then the threshold is a sample-selection artifact rather than a physical separator. A second check is to compare the circularity-based early/late labels against the five-component classifications in TNG50 for the same galaxies; if many galaxies with substantial thin-disk mass are placed in the early-type class by the 0.25 cut, the proxy fails its own validation.

Watch

Extended reading notes

Core claim

The central claim is that the stellar circularity parameter $\varepsilon = j_z/j_c(E)$ — the ratio of a star's angular momentum component along the galaxy's rotation axis to the angular momentum it would have on a circular orbit at the same binding energy — carries enough information for broad morphology classification. Defining the disk fraction $F_{\rm disk}$ as the mass fraction of stars with $\varepsilon > 0.7$ minus the fraction with $\varepsilon < -0.7$, and the spheroid fraction $F_{\rm sph}$ as twice the mass fraction below $\varepsilon = 0$, the paper finds these two scalars map onto the physical components that matter: $F_{\rm disk}$ traces the thin disk, and $F_{\rm sph}$ traces the classical bulge plus stellar halo. A local minimum in the $F_{\rm disk}$ distribution near 0.225 for galaxies with $10 \le \log(M_\star/M_\odot) \le 12$ motivates a round threshold of 0.25, lower than the 0.4 used in earlier work, and systems with $F_{\rm disk}+F_{\rm sph}<2/3$ are called irregular or morphologically complex. Applied to $z=0$ galaxies in TNG100, the classification makes late-type galaxies the majority in the field and early types the majority in clusters, which the paper reads as tentative evidence that the morphology-density relation is reproduced.

Load-bearing premise

The load-bearing premise is that the two-peaked shape of the disk-fraction distribution that fixes the 0.25 threshold is a stable property of galaxies with $10 \le \log(M_\star/M_\odot) \le 12$, so the same cutoff can be applied to lower-mass galaxies down to $\log(M_\star/M_\odot)=9$ and across all environments, while the cutoff identifying irregular systems is chosen by hand rather than derived.

Editorial extensions

If this is right

  • Galaxy morphology labels can be assigned from precomputed catalog quantities alone, with no need to run a five-component orbit decomposition or construct mock images.
  • A threshold of 0.25 in the disk fraction, lower than the commonly used 0.4, keeps moderately disk-dominated systems — plausible lenticulars and dynamically heated spirals — inside the late-type class.
  • Because the circularity catalog spans every simulation box and snapshot, the same classification can be applied across mass, volume, and redshift for large statistical samples.
  • The recovery of a morphology-density relation in TNG100, even with a coarse field/group/cluster split, shows that the disk fraction itself carries environmental information.
  • The inexpensive labels are well suited to serve as a baseline or training set for machine-learning morphology classifiers.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If the 0.25 threshold is stable, the same decomposition can be pushed to higher redshift, where visual morphology is harder to measure; the all-snapshot coverage makes it straightforward to test whether the two-peaked distribution and its minimum persist beyond $z=0$.
  • The paper's environmental result does not identify the mechanism behind the morphology-density relation; a natural extension is to check whether the relation survives within fixed stellar-mass bins or after splitting by quiescence, separating environmental disk removal from mass-driven transformation.
  • The 2/3 cutoff for irregular and complex systems is a hand-chosen parameter, so an immediate refinement is to tune it against the detailed five-component labels in TNG50, turning an arbitrary boundary into a measured one.
  • Because disk fraction correlates with gas content and star formation, the same classification could act as a bridge between simulated kinematics and photometric morphology in observational surveys.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 4 minor

Summary. The paper investigates whether the stellar orbital circularity parameters in the IllustrisTNG public 'Stellar Circularities, Angular Momenta, Axis Ratios' catalog (catalog c) can serve as a simple morphological proxy. The authors first validate catalog (c) against the more detailed five-component kinematic decomposition in TNG50 (catalog t), finding that the circularity-based disk fraction correlates most strongly with the thin-disk component and that the circularity-based 'bulge' corresponds to the sum of classical bulge and stellar halo. They then apply this decomposition to TNG100 at z=0 and propose a threshold of Fdisk=0.25, derived from a bimodal Fdisk distribution in the stellar-mass range 10≤log(M*/M_sun)≤12, to separate early- and late-type galaxies, with an additional hand-set cut (Fdisk+Fsph<2/3) to remove irregular/complex systems. Using this classification on the mass range 9≤log(M*/M_sun)≤12, they report a morphology-density trend with late-types dominant in the field and early-types dominant in clusters, concluding that circularity alone is a viable and efficient morphological proxy.

Significance. If the central claims are established, this would be a useful and inexpensive tool: catalog (c) already exists for all TNG boxes and all snapshots, so a validated circularity-based classification would enable large statistical studies of morphology and a baseline/training set for machine learning. The TNG50 validation in §2.1 is a genuine strength: it uses an independent, more detailed decomposition and gives the physical meaning of the catalog (c) parameters. However, the paper's central threshold has not been independently validated for the mass range to which it is applied, and the morphology-density result is a consistency check of the same classification rather than an external validation. The overstatement in the conclusion relative to the tentative body text also needs correction.

major comments (4)
  1. [§3 and §3.1, Figs. 6 and 8, Table 1] The threshold Fdisk=0.25 is derived from the bimodal Fdisk distribution in the restricted range 10≤log(M*/M_sun)≤12 (Fig. 6), yet the classification is applied to the environmental sample with 9≤log(M*/M_sun)≤12. The paper itself shows (Figs. 4 and 5) that for lower masses the distribution is not bimodal and is dominated by low-Fdisk galaxies, so the justification for the threshold does not extend to the 9–10 mass bin. Please restrict the environmental analysis to the mass range used for threshold selection, or demonstrate that the threshold and the bimodal minimum are stable across the full mass range (e.g., by repeating the procedure in narrower mass bins).
  2. [§3, text around Fig. 6] The irregular/complex cut Fdisk+Fsph<2/3 is introduced with qualitative reasoning but no sensitivity analysis. Because this cut removes galaxies before the Fdisk distribution in Fig. 6 is computed, the adopted 2/3 value directly affects the location of the reported minimum at Fdisk≈0.225 and hence the chosen 0.25 threshold. Please show how varying the irregularity threshold changes (i) the main classification threshold and (ii) the environmental fractions in Table 1, and provide a quantitative or empirically justified criterion for the 2/3 choice.
  3. [§2.1 and §3.1] The validation in §2.1 establishes a correlation between Fdisk and the thin-disk fraction from the TNG50 catalog (t), but a correlation does not test whether a binary split at Fdisk=0.25 corresponds to true early- versus late-type galaxies. The morphology-density result in Table 1 is not independent evidence, because it is produced by the same threshold-based classification being assessed. To support the abstract's claim that circularities are 'enough,' the paper should test the threshold against an independent morphological classification in TNG100 (e.g., the mock-image classifications of Rodriguez-Gomez et al. 2019 or the kinematic decomposition of Du et al. 2019, 2020), or at minimum report the completeness and purity of the Fdisk=0.25 split using the TNG50 catalog (t) labels.
  4. [§5 vs §4] The conclusion says the results 'demonstrate' and 'support the reliability' of the classification, while the body repeatedly describes the evidence as 'tentative' (§3.1 and §4). Given the unvalidated threshold and the hand-set irregularity cut, the conclusion overstates the strength of the evidence. Either temper the conclusion to match the tentative framing or add the validation needed to support the stronger claim.
minor comments (4)
  1. [Fig. 1] The correlation matrix would be clearer if the axes explicitly identified which catalog (c) parameter ('CircAbove07MinusBelow07Frac' or 'CircTwiceBelow0Frac') is meant by 'Disk (c)' and 'Bulge (c)', and if the within-10-times-half-mass-radius choice was stated in the caption.
  2. [§3.1, Table 1] Providing fractions or percentages alongside the raw counts would make the morphology-density comparison across environments more direct, especially since the totals differ substantially (2983, 9921, 8133).
  3. [References] There is a formatting artifact in the reference list: ' Lokas, E. L.' has a leading space; please check the bibliography style consistency throughout.
  4. [Title page and Serbian summary] The manuscript includes journal template artifacts at the top (e.g., 'Serb. Astron. J. } 200 (2020), 1 - 5 UDC 52') and the Serbian summary is placed after the English references; these should be cleaned up for submission.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the Fdisk threshold is a data-informed calibration choice, and the morphology-density test uses independently defined environments.

full rationale

The paper's derivation chain is self-contained with respect to circularity. The primary proxy Fdisk is an externally defined catalog quantity (CircAbove07MinusBelow07Frac from Genel et al. 2015), and its physical interpretation is checked against the independent five-component Mordor decomposition of Zana et al. (2022) in TNG50. The classification threshold Fdisk = 0.25 is chosen from the observed bimodality of the TNG100 Fdisk distribution, so the early/late labels are by construction functions of the same variable. However, the paper does not present this as a prediction; it is an explicit calibration step. The subsequent morphology-density analysis uses environments defined from Friends-of-Friends halo masses and subhalo counts, which are independent of Fdisk, so the recovered environmental trend is not forced by the threshold choice. The hand-set irregular/complex cut (Fdisk + Fsph < 2/3) and the absence of a sensitivity analysis are robustness limitations, not circularity. No load-bearing self-citations, imported uniqueness theorems, or ansatz-smuggling are present. The authors' conclusions are appropriately hedged as preliminary and tentative.

Assumptions & free parameters 4 free parameters · 5 assumptions · 0 invented entities

Four numbers are chosen from the data rather than derived: the disk-fraction split, the irregular-galaxy cutoff, and two stellar-mass windows. The first three directly shape the classification; the mass window for the environmental test alters the reported morphology-density counts. None of these is fitted to an external observational target, so the circularity burden remains modest.

free parameters (4)
  • Fdisk classification threshold = 0.25
    Selected as a value slightly above the local minimum (0.225) of the Fdisk distribution for galaxies with 10 ≤ log(M*/M_sun) ≤ 12 (Fig 6). The choice drives the early/late classification.
  • Irregular/complex threshold (Fdisk+Fsph) = 2/3
    Chosen by hand to isolate galaxies whose disk plus spheroid fractions are far below unity; no sensitivity analysis or external calibration is provided.
  • Stellar-mass range for threshold selection = 10 ≤ log(M*/M_sun) ≤ 12
    The bimodal shape used to motivate Fdisk=0.25 appears only in this restricted range; the full sample distribution is monotonically decreasing (Fig 4).
  • Stellar-mass range for environmental analysis = 9 ≤ log(M*/M_sun) ≤ 12
    Adopted to reduce contamination from low-mass early-type galaxies; affects the counts in Table 1 and the recovered morphology-density trend.
assumptions (5)
  • domain assumption IllustrisTNG simulations provide a realistic population of galaxies at z=0.
    The entire analysis assumes simulated galaxies represent real galaxies well enough for morphological classification to be meaningful (Section 2).
  • domain assumption The circularity catalog (c) parameters CircAbove07MinusBelow07Frac and CircTwiceBelow0Frac are correct measures of thin disk and spheroid mass fractions.
    The paper's validation against TNG50 supports this, but the raw definitions come from Genel et al. (2015) and are not re-derived.
  • domain assumption The Mordor five-component decomposition (catalog (t)) is an accurate reference standard.
    Used as ground truth for validation; accuracy of Mordor is taken from Zana et al. (2022) rather than independently checked.
  • ad hoc to paper The bimodal Fdisk distribution in the 10-12 mass range reflects a genuine dichotomy between early- and late-type galaxies.
    This interpretation motivates the threshold Fdisk=0.25 but is not independently justified; the full-sample distribution is unimodal.
  • domain assumption The three-tier environment classification (field, group, cluster) captures the physical density relevant to the morphology-density relation.
    Based on Paul et al. (2017) halo mass and subhalo count criteria; the paper acknowledges it is simpler than local density metrics (Section 3.1).

how reviews work

0 comments
Cite this review

Pith. "Pith review of Galaxy Morphology Classification: Are Stellar Circularities Enough?." pith.science (2026). https://pith.science/paper/RTDTJL2U

@misc{pith2026250620002,
  author       = {Pith},
  title        = {Pith review of: Galaxy Morphology Classification: Are Stellar Circularities Enough?},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/RTDTJL2U}},
  note         = {Machine review of arXiv:2506.20002}
}
abstract

We present a preliminary study exploring whether the stellar orbital circularity of simulated galaxies, available from precomputed catalogs in the IllustrisTNG project, can be used as a proxy for broad morphological classification. We focus on the publicly available "Stellar Circularities, Angular Momenta, Axis Ratios" catalog, which enables a simple kinematic decomposition of the stellar component into disk and spheroid subsystems. By validating this approach against the more detailed five-component kinematic decomposition in TNG50, we confirm that the circularity-based disk fraction correlates most strongly with the thin disk, while the bulge fraction broadly represents the combined contribution of classical bulges and stellar halos. We then apply this decomposition to galaxies in the TNG100 simulation at redshift $z=0$ and identify a data-motivated threshold of $\mathrm{F_{disk}} = 0.25$ to distinguish early- and late-type galaxies. This threshold, lower than the commonly adopted value of $0.4$, better captures the diversity of disk-dominated systems and avoids excluding galaxies with moderately prominent disks. Additionally, we identify irregular or morphologically complex systems based on galaxies with low total disk and spheroid mass fractions. Using this classification, we recover a morphology-density relation that is broadly consistent with observations: late-type galaxies dominate in the field, while early-type galaxies are the most prevalent morphological type in clusters. Our results demonstrate that stellar circularity alone can serve as an accessible and computationally efficient morphological proxy. We also discuss the potential for this classification to support machine learning efforts as a baseline or training set for future morphological studies.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

37 extracted references · 25 canonical work pages

  1. [1]

    Blanton, M. R. and Moustakas, J. 2009, ARA&A, 47, 159

    work page 2009

  2. [2]

    Conselice, C. J. 2014, ARA&A, 52, 291

    2014

  3. [3]

    Crain, R. A. and van de Voort, F. 2023, ARA&A, 61, 473 de Vaucouleurs, G. 1959, Handbuch der Physik, 53, 275

    work page 2023

  4. [4]

    C., Debattista, V

    Du, M., Ho, L. C., Debattista, V. P., et al. 2020, ApJ, 895, 139

    work page 2020

  5. [5]

    C., Zhao, D., et al

    Du, M., Ho, L. C., Zhao, D., et al. 2019, ApJ, 884, 129 Falc´ on-Barroso, J. and Knapen, J. H. 2013, Secular Evo- lution of Galaxies (Cambridge University Press)

    work page 2019

  6. [6]

    A., Jaque Arancibia, M., Dolfi, A., and Monsalves, N

    Fontirroig, V., Gomez, F. A., Jaque Arancibia, M., Dolfi, A., and Monsalves, N. 2024, arXiv e-prints, in press (DOI: 10.48550/arXiv.2411.19723), arXiv:2411.19723 Gal´ an-de Anta, P. M., Sarzi, M., Pillepich, A., et al. 2022, MNRAS, 517, 5992

    arXiv doi:10.48550/arxiv.2411.19723 2024

  7. [7]

    M., Hernquist, L., et al

    Genel, S., Fall, S. M., Hernquist, L., et al. 2015, ApJL, 804, L40

    2015

  8. [8]

    2025, arXiv e-prints, in press (DOI: 10.48550/arXiv.2504.18042), arXiv:2504.18042

    Gong, J.-Y., Lin, W., Tang, L., and Lan, Y. 2025, arXiv e-prints, in press (DOI: 10.48550/arXiv.2504.18042), arXiv:2504.18042

    arXiv doi:10.48550/arxiv.2504.18042 2025

Show all 37 references
  1. [9]

    R., Millman, K

    Harris, C. R., Millman, K. J., van der Walt, S. J., et al. 2020, Natur, 585, 357

    2020

  2. [10]

    Hubble, E. P. 1926, ApJ, 64, 321

    1926

  3. [11]

    2019, MNRAS, 489, 1859

    Huertas-Company, M., Rodriguez-Gomez, V., Nelson, D., et al. 2019, MNRAS, 489, 1859

    2019

  4. [12]

    Hunter, J. D. 2007, Computing in Science & Engineering, 9, 90

    2007

  5. [13]

    D., Pillepich, A., Nelson, D., et al

    Joshi, G. D., Pillepich, A., Nelson, D., et al. 2020, MN- RAS, 496, 2673

    2020

  6. [14]

    and Kennicutt, Robert C., J

    Kormendy, J. and Kennicutt, Robert C., J. 2004, ARA&A, 42, 603 Lokas, E. L. 2022, A&A, 662, A53

    2004

  7. [15]

    2018, MNRAS, 480, 5113

    Marinacci, F., Vogelsberger, M., Pakmor, R., et al. 2018, MNRAS, 480, 5113

    2018

  8. [16]

    A., Morgan, W

    Matthews, T. A., Morgan, W. W., and Schmidt, M. 1964, ApJ, 140, 35

    1964

  9. [17]

    C., and White, S

    Mo, H., van den Bosch, F. C., and White, S. 2010, Galaxy Formation and Evolution (Cambridge Univer- sity Press)

    2010

  10. [18]

    V., Dalal, N., and Gottl¨ ober, S

    More, S., Kravtsov, A. V., Dalal, N., and Gottl¨ ober, S. 2011, ApJS, 195, 4

    2011

  11. [19]

    Morgan, W. W. and Mayall, N. U. 1957, PASP, 69, 291

    1957

  12. [20]

    Morgan, W. W. and Osterbrock, D. E. 1969, AJ, 74, 515

    1969

  13. [21]

    P., Pillepich, A., Springel, V., et al

    Naiman, J. P., Pillepich, A., Springel, V., et al. 2018, MNRAS, 477, 1206

    2018

  14. [22]

    2018, MN- RAS, 475, 624

    Nelson, D., Pillepich, A., Springel, V., et al. 2018, MN- RAS, 475, 624

    2018

  15. [23]

    J., et al

    Park, M., Tacchella, S., Nelson, E. J., et al. 2022, MN- RAS, 515, 213

    2022

  16. [24]

    S., Gupta, P., and Kumar, H

    Paul, S., John, R. S., Gupta, P., and Kumar, H. 2017, MNRAS, 471, 2

    2017

  17. [25]

    K., Bekki, K., et al

    Pfeffer, J., Cavanagh, M. K., Bekki, K., et al. 2023, MN- RAS, 518, 5260

    2023

  18. [26]

    2019, MN- RAS, 490, 3196

    Pillepich, A., Nelson, D., Springel, V., et al. 2019, MN- RAS, 490, 3196

    2019

  19. [27]

    2018b, MN- RAS, 473, 4077 Planck Collaboration, Ade, P

    Pillepich, A., Springel, V., Nelson, D., et al. 2018b, MN- RAS, 473, 4077 Planck Collaboration, Ade, P. A. R., Aghanim, N., et al. 2016, A&A, 594, A13

    2016

  20. [28]

    F., Lotz, J

    Rodriguez-Gomez, V., Snyder, G. F., Lotz, J. M., et al. 2019, MNRAS, 483, 4140

    2019

  21. [29]

    Sellwood, J. A. 2014, Reviews of Modern Physics, 86, 1

    2014

  22. [30]

    2010, MNRAS, 401, 791

    Springel, V. 2010, MNRAS, 401, 791

    2010

  23. [31]

    and Hernquist, L

    Springel, V. and Hernquist, L. 2003, MNRAS, 339, 289

    2003

  24. [32]

    2018, MN- RAS, 475, 676

    Springel, V., Pakmor, R., Pillepich, A., et al. 2018, MN- RAS, 475, 676

    2018

  25. [33]

    2022, MNRAS, 509, 2654

    Varma, S., Huertas-Company, M., Pillepich, A., et al. 2022, MNRAS, 509, 2654

    2022

  26. [34]

    2020, Nature Reviews Physics, 2, 42

    Vogelsberger, M., Marinacci, F., Torrey, P., and Puch- wein, E. 2020, Nature Reviews Physics, 2, 42

    2020

  27. [35]

    Waskom, M. L. 2021, Journal of Open Source Software, 6, 3021

    2021

  28. [36]

    2017, MNRAS, 465, 3291 Wes McKinney

    Weinberger, R., Springel, V., Hernquist, L., et al. 2017, MNRAS, 465, 3291 Wes McKinney. 2010, in Proceedings of the 9th Python in Science Conference, ed. St´ efan van der Walt and Jarrod Millman, 56 – 61

    2017

  29. [37]

    Zana, T., Lupi, A., Bonetti, M., et al. 2022, MNRAS, 515, 1524 10 Galaxy Morphology Classification: Are Stellar Circularities Enough? MORFOLOXKA KLASIFIKACIJA G ALAKSIJA: DA LI SU ZVEZDANE CIRKULARNOSTI DOVOLjNE? Katarina Baucalo1 and Ana Mitraˇ sinovi´ c2 1Department of Astron...

    2022

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.