Pith. sign in

REVIEW 4 major objections 5 minor 250 references

The globular cluster M 22 contains at least two stellar populations distinguished by a real iron abundance spread of ≥0.24 dex and an average s-process element spread of ~0.65 dex, and the spread persists within each population.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

T0 review · deepseek-v4-flash

2026-08-02 02:52 UTC pith:YN2W3XPL

load-bearing objection High-quality differential analysis that likely confirms M22's Fe and s-process spreads, though the six-star sample and unclosed 1D LTE systematics keep the result short of definitive. the 4 major comments →

arxiv 2607.14065 v1 pith:YN2W3XPL submitted 2026-07-15 astro-ph.GA

The complex stellar system M 22: confirming abundance variations with high precision differential measurements

M. McKenzie , D. Yong , A. F. Marino , S. Monty , E. Wang , A. I. Karakas , A. P. Milone , M. V. Legnardi
show 3 more authors
I. U. Roederer S. Martell D. Horta
This is my paper
classification astro-ph.GA
keywords M 22NGC 6656globular clustersstellar abundancesmultiple stellar populationss-process elementsdifferential spectroscopyType II clusters
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

The paper tries to settle a long dispute over whether M 22, a globular cluster-like system, hosts real, intrinsic variations in heavy-element abundances. Using six red giant stars observed at very high spectral resolution and signal-to-noise, the authors apply a strictly differential, star-by-star abundance analysis that reaches uncertainties near 0.01 dex—about 2 percent. They find an iron spread of at least 0.24 dex between two stellar populations and an average slow-neutron-capture (s-process) element spread of about 0.65 dex, and they show that each population has its own internal abundance variations, with the metal-poor population showing larger spreads in elements heavier than iron. If correct, this makes M 22 a bona fide 'Type II' globular cluster—one of a small class of systems with heavy-element variations—and rules out the idea that earlier detections were artifacts of low precision or reddening. The result matters because it constrains how M 22 formed: as a stripped galactic nucleus, the product of two merged clusters, or a surviving building block of the Milky Way's early disc.

Core claim

Differential line-by-line analysis of very high quality (R = 110,000, signal-to-noise ≈ 300) spectra of six red giant branch stars in M 22 confirms that the cluster is chemically inhomogeneous. The data show a clear split into two populations: a metal-rich group enhanced in s-process elements and a metal-poor group that is not, with a minimum iron spread of 0.24 dex and an average s-process spread of 0.65 dex. Abundance variations are detected in 18 elements—Fe, Na, Si, Ca, Sc, Ti, Cr, Mn, Co, Ni, Zn, Y, Zr, La, Ce, Nd, Sm, and Eu. The variations persist, and are in some cases larger, within each individual population, especially among the metal-poor stars. The authors interpret this as evid

What carries the argument

The key machinery is the strictly differential, line-by-line abundance analysis. Each program star's spectrum is measured against the same reference giant (NGC 6752-mg9) using the same line list and the same one-dimensional, local-thermodynamic-equilibrium model-atmosphere grid, so that errors in atomic data, model atmospheres, and continuum placement cancel to first order in every line. Equivalent widths are measured with a purpose-built fitting routine that combines Gaussian fits with a neural network trained on synthetic spectra, and strong or weak lines that would introduce systematic errors are removed. This reduces relative abundance uncertainties to roughly 0.01 dex and lets the autho

Load-bearing premise

All of the claimed spreads rest on the assumption that one-dimensional, local-thermodynamic-equilibrium model atmospheres and the differences between the reference star and the M 22 giants introduce systematic errors far smaller than the quoted 0.01–0.02 dex uncertainties; if population-dependent systematics, such as CNO-induced changes to the model structure, are larger than the estimated ~0.001 dex differential NLTE corrections, the reported iron and s-process spreads could

What would settle it

Re-derive iron and s-process abundances for the same six stars using a full 3D non-LTE pipeline with photometric effective temperatures and literature C+N+O abundances. If the differential NLTE correction between the most metal-poor star and the reference giant exceeds about 0.02 dex, the ≥0.24 dex iron spread is not yet established; if the correction stays at the ~0.001 dex level estimated in the paper, the spread is confirmed. A complementary larger-sample test: obtain the same quality spectra for 20+ additional RGB stars across the s-process divide; if the bimodal separation does not recur,

Watch this falsifier — get emailed when new claim-graph text bears on it.

If this is right

  • M 22 joins the small class of Type II globular clusters, like ω Centauri and Terzan 5, that show real metallicity and/or s-process variations, so any model of Galactic globular cluster formation must explain why a disc cluster has this property.
  • The iron spread and its persistence when different reference stars and stellar parameter choices are used mean that earlier null detections most likely resulted from lower precision or from neglecting C+N+O abundance differences, not from the absence of a spread.
  • The within-population spreads in heavy elements, which grow with atomic number, argue that the gas from which each population formed was itself not chemically uniform, favouring enrichment from multiple nucleosynthetic sources rather than a single pollution event.
  • The strong correlation between Si and Y abundance differences and the overall chemical patterns provide direct constraints on merger, nuclear-star-cluster, and early-disc formation scenarios for M 22.
  • A kinematic analysis combining M 22's disc-like orbit with these chemical data suggests that if M 22 is a stripped nucleus, the merger that brought it in must have been gentle and at low inclination; if it is a merged pair of clusters, the merger must have occurred early enough for the heavy-element pattern to be imprinted.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The paper's differential analysis hinges on the assumption that the reference giant and the M 22 stars have identical line-formation physics. If the C+N+O abundance difference between the two M 22 populations, estimated at about 0.13 dex, changes the model atmosphere temperature structure in a way the 1D grid does not capture, the quoted 0.24 dex iron spread could be inflated by a few hundredths o
  • The increasing heavy-element spread with atomic number within the metal-poor population, if confirmed by a larger sample, would suggest that the s-process yield in the cluster's early gas depended on neutron exposure, i.e., on the mass of the polluting AGB stars. This could be tested by comparing the observed Y, Zr, La, Ce, Nd, and Sm patterns to nucleosynthesis yields from AGB models of 3–6 solar
  • If M 22 really is a surviving building block of the Milky Way's disc, then other, fainter disc clusters or dissolved remnants with a similar bimodal s-process pattern should exist. Wide-field, high-resolution surveys of metal-poor disc stars could search for stars with high [La/Eu] at [Fe/H] ≈ -1.7 and disc-like orbits; finding such a population would strengthen the building-block scenario.
  • Since the six stars were selected to have detectable MgH lines for future Mg isotope analysis, the same data could be used to measure Mg isotopic ratios. If the two s-process populations differ in 25Mg/24Mg and 26Mg/24Mg, that would favour AGB self-enrichment; if the ratios are identical, the two populations would more likely have formed from gas enriched by different external sources.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 5 minor

Summary. The paper presents a strictly differential line-by-line abundance analysis of six red giant branch stars in M 22, using high-resolution (R = 110,000), high-S/N (≈300) VLT/UVES spectra and the q2/MOOG pipeline with NGC 6752-mg9 as the reference star. The authors report an iron abundance spread of at least 0.24 dex, a separation between two s-process populations with an average s-process spread of about 0.65 dex, and non-negligible abundance variations within each population. They argue that these results confirm M 22 as a Type II cluster and discuss three possible origins: a nuclear star cluster, a merger product, or a building block of the Milky Way.

Significance. If the central claim is robust, this is an important result: it would settle a long-standing debate about whether M 22 hosts intrinsic heavy-element abundance variations, strengthen the Type II cluster classification, and constrain scenarios for its formation. The paper has clear strengths: the spectra are of exceptional quality, the differential approach uses ~100 Fe I lines per star, internal errors reach ~0.01 dex, and the authors test the sensitivity of the Fe spread to different reference stars and initial parameters (Fig. 8). The public release of the REvIEW code is also a positive feature. The main weakness is that the systematic uncertainties from 1D LTE model atmospheres, CNO abundance variations, and 3D/NLTE effects are not quantified for the elements that drive the main claims.

major comments (4)
  1. [§2.3 and §4.1.1] The central Fe-spread and s-process-spread claims rest on differential 1D LTE ATLAS9 models with [α/Fe]=+0.4 and no CNO enhancement. M 22 populations are reported to differ by Δ[C+N+O]≈0.13 dex (§4.1.3), which can alter the T-τ structure and line formation for both Fe I/Fe II and s-process lines. The paper estimates differential NLTE corrections only for Fe (INSPECT, §4.1.1), not for La II, Ce II, Nd II, or Eu II, and does not test CNO-enhanced model atmospheres or 3D corrections. Given that the observed ΔFeI range is 0.235 dex and the within-population σ values are 0.04–0.06 dex, a population-dependent systematic of ~0.05 dex would be comparable to the signal. I request that the authors quantify such systematics or provide explicit model tests with CNO-enhanced and/or 3D atmospheres.
  2. [Abstract and §3.1] The abstract states an "average s-process abundance spread of 0.65 dex", but §3.1 reports an average separation of ≈0.5 dex between the two s-process groups and the largest element spread (Y) as "almost 0.6 dex". No table or figure in the paper supports 0.65 dex. This quantitative inconsistency needs to be reconciled or corrected before publication.
  3. [§3.2, Tables 4–5] Several of the elements used to support the claim of intra-population abundance variations are measured from one line per star (Eu, Zn, Zr, Sm, and Cr II in some cases). The authors themselves note that the low Eu abundance of IV-102, which drives the s-poor population Eu dispersion, "may be an artefact of our differential approach" (§3.2). Because the within-population dispersion is a distinct headline claim (larger heavy-element spreads in the metal-poor population), the paper should explicitly separate results based on multiple lines from those based on single lines and quantify how a single-line systematic would affect the conclusion.
  4. [§3.1] The sample is six stars selected from Marino et al. (2011) based on their s-process group and detectability of MgH lines, with only three stars per population. The authors note that "results are limited by our small sample size", but the abstract and conclusions present cluster-wide statements about the spread and its population dependence. The intrinsic-spread claim per se is supported by the differential precision, but the distribution-level statements (e.g., "the more metal-poor population hosting larger spreads") need to be framed with appropriate caveats about selection and sample size.
minor comments (5)
  1. [Table 4] Column header "⟨s−rich⟩−⟨s−rich⟩" appears to be a typo; it should read "⟨s−poor⟩−⟨s−rich⟩".
  2. [Appendix B] The text notes that the hyperparameter selection uses a labelled real data set, which "does cause data leakage". This is transparent, but the potential impact on the FFNN validation should be discussed more explicitly, since the code is a component of the analysis.
  3. [References] Yong et al. 2014a and 2014b are listed with identical journal, volume, and page numbers (MNRAS, 441, 3396). This is likely a bibliographic error and should be corrected.
  4. [Fig. 8] The caption states that abundances have been scaled by the reference metallicity but also says "[Fe/H] abundances should not be compared between methods". The scaling makes the figure difficult to interpret; a clearer explanation of what is plotted is needed.
  5. [Table C1] The "This work" [Fe/H] values are relative to NGC 6752-mg9, while the literature values are absolute. This should be labeled explicitly in the table to avoid confusion.

Circularity Check

1 steps flagged

Minor inherited grouping from Marino et al. (2011), but the differential abundance measurements and reference-star robustness tests make the central claim largely self-contained.

specific steps
  1. self definitional [Sections 2.1 and 3.1; Figs 1 and 4]
    "We select our program stars from Marino et al. (2011). ... Three of which belong to the s-process poor group as identified by Marino et al. (2011), and the remaining three are from the s-process rich population. ... We reproduce the divide between the two s-process groups using ΔLa−Eu as a function of ΔFe... As in Marino et al. (2011), we see a clear separation in our program stars and find that there must be at least a 0.24 dex iron abundance spread in the cluster as well as a spread of at least 0.38 dex for ΔLa−Eu."

    The two stellar populations are not inferred from the new sample: the targets were deliberately drawn from Marino et al.'s pre-existing s-rich and s-poor groups, which are defined by the same [La/Eu] ratio used to quantify the s-process spread. The reported 0.65 dex s-process spread and ΔLa−Eu separation in Fig. 4 are therefore statistics of a sample constructed to contain that separation; the new EW measurements can validate the prior values but cannot independently establish the bimodality. Because the central Fe-spread claim also comes from these pre-selected groups (C vs IV-102), the 'confirmation' is partly inherited from the input classification, though the differential measurements and robustness tests provide independent content.

full rationale

The paper's central result is a differential abundance measurement, not a first-principles derivation, so most circularity patterns do not apply. q2 uses Fe excitation/ionization balance to set stellar parameters and then reports Fe abundances; this is standard and the paper tests robustness by varying reference stars and initial parameters (Fig. 8), including the independent Gaia benchmark HD 122563. The NLTE check uses the external INSPECT database. The paper also openly notes its small sample size and the possible IV-102 Eu artefact. The only substantive caveat is the inherited s-rich/s-poor grouping from Marino et al. (2011), which makes the two-population claim a confirmation of a pre-selected split rather than an unbiased detection. This is a selection limitation, but the measured abundances and error budget are new, and the Fe spread persists across reference-star changes, so the central claim does not reduce by construction to a fit or to a self-citation chain.

Axiom & Free-Parameter Ledger

8 free parameters · 6 axioms · 0 invented entities

The central claim relies on a standard differential spectroscopic analysis with 6 sets of stellar parameters (24 fitted values), two EW cutoffs, and several model assumptions. No new physical entities are introduced. The free parameters are not arbitrary fitting knobs—they are constrained by the spectra—but they are nevertheless fitted quantities and the result depends on their reliability.

free parameters (8)
  • Stellar parameters for star C (Teff, log g, [Fe/H], xi) = 3912 K, 0.105, -1.696, 2.08 km/s
    Determined by q2 via excitation/ionization balance with Fe I/Fe II; these parameters set the abundance scale and thus directly affect the measured Fe and element spreads.
  • Stellar parameters for star III-3 (Teff, log g, [Fe/H], xi) = 4041 K, 0.250, -1.778, 2.29 km/s
    Same as above; fitted to achieve balance.
  • Stellar parameters for star III-14 (Teff, log g, [Fe/H], xi) = 4038 K, 0.120, -1.870, 2.24 km/s
    Same as above.
  • Stellar parameters for star III-15 (Teff, log g, [Fe/H], xi) = 4136 K, 0.450, -1.825, 2.03 km/s
    Same as above.
  • Stellar parameters for star III-52 (Teff, log g, [Fe/H], xi) = 4100 K, 0.510, -1.707, 1.93 km/s
    Same as above.
  • Stellar parameters for star IV-102 (Teff, log g, [Fe/H], xi) = 4043 K, 0.100, -1.973, 2.43 km/s
    Same as above.
  • Strong-line EW cutoff = 120 mÅ
    Lines above this threshold are excluded because their EWs are insensitive to microturbulence; a hand-chosen modelling threshold affecting line selection.
  • Weak-line EW cutoff = 5 mÅ
    Lines below this threshold are excluded due to larger fractional uncertainties; a hand-chosen cutoff.
axioms (6)
  • domain assumption 1D LTE plane-parallel ATLAS9 model atmospheres with [alpha/Fe]=+0.4 are adequate for the program stars, and differential analysis cancels residual model errors.
    Section 2.3 states the grid used; the analysis relies on this cancellation for the claimed precision.
  • domain assumption Reference star NGC 6752-mg9 has accurate photometric parameters (Teff=4288K, logg=0.91, [Fe/H]=-1.66, xi=1.72) and is a valid differential anchor for M22 RGB stars.
    Section 2.3 / Table 3; the abundance differences are measured with respect to this star, and parameter errors propagate into the reported uncertainties.
  • domain assumption Excitation and ionization balance of Fe I and Fe II yields unbiased relative stellar parameters for metal-poor giants.
    Section 2.3; q2 imposes these balances on the differential abundances. If Fe II is biased by NLTE, the derived log g and the final abundances could be affected.
  • domain assumption The [La/Eu] separation criterion from Marino et al. (2011) identifies two genuine stellar populations in M22.
    Section 3.1; the paper uses this prior classification to assign stars to s-rich and s-poor groups, and the 'two populations' claim depends on that grouping being real.
  • domain assumption Equivalent widths measured by REvIEW are accurate to ~1–2 mÅ; comparisons with DAOSPEC and IRAF show small offsets.
    Section 2.2 / Fig. 3; any systematic error in EW measurement that varies between the s-rich and s-poor stars would directly fake abundance spreads.
  • domain assumption The stellar lines used are unblended (or properly deblended) and the continuum placement is correct.
    Section 2.2 / Appendix B; manual inspection is described, but not quantitatively verified for every line. Blends could bias individual line abundances.

pith-pipeline@v1.3.0-alltime-deepseek · 28137 in / 13844 out tokens · 131628 ms · 2026-08-02T02:52:15.979001+00:00 · methodology

0 comments
read the original abstract

M 22 (NGC 6656) is a chemically complex globular cluster-like system reported to harbour heavy element abundance variations. However, the extent of these variations and the origin of this cluster is still debated. In this work, we investigate the chemical inhomogeneity of M 22 using differential line-by-line analysis of high-quality (R = 110,000, S/N = 300 per pixel at 514 nm) VLT/UVES spectra of six carefully chosen red giant branch stars. By achieving abundance uncertainties as low as ~0.01 dex (~2 per cent), this high-precision data validates the results of previous studies and reveals variations in Fe, Na, Si, Ca, Sc, Ti, Cr, Mn, Co, Ni, Zn, Y, Zr, La, Ce, Nd, Sm and Eu. Additionally, we can confirm that the cluster hosts two stellar populations with a spread of at least 0.24 dex in [Fe/H] and an average s-process abundance spread of 0.65 dex. In addition to global variations across the cluster, we also find non-negligible variations within each of the two populations, with the more metal-poor population hosting larger spreads in elements heavier than Fe than the metal-rich. We address previous works which do not identify anomalous abundances and relate our findings to our current dynamical understanding of the cluster. Given our results, we suggest that M 22 is either a nuclear star cluster, the product of two merged clusters, or an original building block of the Milky Way.

Figures

Figures reproduced from arXiv: 2607.14065 by A. F. Marino, A. I. Karakas, A. P. Milone, D. Horta, D. Yong, E. Wang, I. U. Roederer, M. McKenzie, M. V. Legnardi, S. Martell, S. Monty.

Figure 1
Figure 1. Figure 1: In each panel we show the our s-process poor stars in blue crosses (III-15, III-14, IV-102) and s-process rich stars in red pentagons (C, III-52, III-3). Left panel: The RGB in M 22 using Gaia filters. All targets appear at the tip of the RGB. Middle panel: The Na-O anti-correlation using the sample of stars from Marino et al. (2011). Our stars span the range of the anti-correlation. There is no distinctio… view at source ↗
Figure 2
Figure 2. Figure 2: High signal-to-noise spectra of our target stars. Spectra increase in metallicity from the 𝑠-process poor stars in blue at the top to the 𝑠-process rich stars in red on the bottom. The blended regions from 5297 Å to 5299 Å are a combination of Cr I, Ti I and Fe I. The s-process element La increases in strength from top to bottom as the metallicity increases [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Comparison of EWs measured using DAOSPEC and with IRAF from Yong et al. (2013). The left panel shows all lines measured (N = 872) using DAOSPEC for our program stars and the right panel compares EW measurements for our reference star (N = 199). Overlaid is a Gaussian fit to the distribution and we find that compared to DAOSPEC, 𝜇 = 0.13 mÅ, 𝜎 = 3.6 mÅ and for Yong et al. (2013); 𝜇 = −0.12 mÅ, 𝜎 = 2.2 mÅ. (… view at source ↗
Figure 4
Figure 4. Figure 4: The Δ La−Eu abundances as a function of Δ Fe (analogous to [La/Eu] and [Fe/H] in square bracket notation respectively). Using high-precision differential abundance measurements we still recover the split between the two s-process populations of stars. The horizontal dashed line in grey represents the Δ La−Eu of the reference star and the horizontal blue and red lines are the average values of Δ La−Eu for t… view at source ↗
Figure 5
Figure 5. Figure 5: Comparing the different abundance variations within each s-process group. 𝛼-elements are shown in green, iron peak elements in yellow,s-process elements in pink and our r-process element, Eu, in purple. Top panel: the 1𝜎 dispersion for the s-process poor group in blue crosses, and the s-process rich group in red pentagons. The s-process poor population has a higher dispersion across all measured s and r-pr… view at source ↗
Figure 6
Figure 6. Figure 6: A grid of correlations between different elements. The Δ X notation is analogous to [X/H] in square bracket notation. On the Y-axis we include Y as a first s-process peak element, Nd as a second s-process peak element and Eu as an r-process element. On the X-axis, we use Si as a proxy for alpha elements and both Fe and Zn as iron peak elements [PITH_FULL_IMAGE:figures/full_fig_p009_6.png] view at source ↗
Figure 8
Figure 8. Figure 8: Each panel illustrates the spread in Fe given different initial values from the literature and different reference stars. The Fe II abundances are shown at the top of every panel, with the same colours and markers used in Figs 4 and 6. For both Fe I and Fe II abundances, the green error bar represents the 1𝜎 standard deviation of Fe for our program stars and the purple box on top of it is the average error… view at source ↗

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

250 extracted references · 34 canonical work pages

  1. [1]

    , keywords =

    Mg isotopic ratios in giant stars of the globular cluster NGC 6752. , keywords =. doi:10.1051/0004-6361:20030296 , archivePrefix =. astro-ph/0303057 , primaryClass =

    Pith/arXiv arXiv

  2. [2]

    , keywords =

    Heavy Elements in Globular Clusters: The Role of Asymptotic Giant Branch Stars. , keywords =. doi:10.1088/0004-637X/785/1/77 , archivePrefix =. 1403.0819 , primaryClass =

    Pith/arXiv arXiv

  3. [3]

    , keywords =

    The s-Process Enrichment of the Globular Clusters M4 and M22. , keywords =. doi:10.1088/0004-637X/795/1/34 , archivePrefix =. 1409.1227 , primaryClass =

    Pith/arXiv arXiv

  4. [5]

    , keywords =

    High precision differential abundance measurements in globular clusters: chemical inhomogeneities in NGC 6752. , keywords =. doi:10.1093/mnras/stt1276 , archivePrefix =. 1307.4486 , primaryClass =

    Pith/arXiv arXiv

  5. [6]

    Two stellar groups with different iron and s-process element abundances

    A double stellar generation in the globular cluster NGC 6656 (M 22). Two stellar groups with different iron and s-process element abundances. , keywords =. doi:10.1051/0004-6361/200911827 , archivePrefix =. 0905.4058 , primaryClass =

    Pith/arXiv arXiv

  6. [7]

    , keywords =

    The two metallicity groups of the globular cluster M 22: a chemical perspective. , keywords =. doi:10.1051/0004-6361/201116546 , archivePrefix =. 1105.1523 , primaryClass =

    Pith/arXiv arXiv

  7. [8]

    arXiv e-prints , keywords =

    How Many Elements Matter?. arXiv e-prints , keywords =

  8. [9]

    Variance and correlation

    Metallicity fluctuation statistics in the interstellar medium and young stars - I. Variance and correlation. , keywords =. doi:10.1093/mnras/stx3286 , archivePrefix =. 1708.06853 , primaryClass =

    Pith/arXiv arXiv

  9. [10]

    , keywords =

    SLUG IV: a novel forward-modelling method to derive the demographics of star clusters. , keywords =. doi:10.1093/mnras/sty2896 , archivePrefix =. 1810.10173 , primaryClass =

    Pith/arXiv arXiv

  10. [11]

    A suite of tools for simulated photometry, spectroscopy, and Bayesian inference with stochastic stellar populations

    SLUG - stochastically lighting up galaxies - III. A suite of tools for simulated photometry, spectroscopy, and Bayesian inference with stochastic stellar populations. , keywords =. doi:10.1093/mnras/stv1374 , archivePrefix =. 1502.05408 , primaryClass =

    Pith/arXiv arXiv

  11. [12]

    Quantifying the effects of stochasticity on star formation rate indicators

    SLUG - Stochastically Lighting Up Galaxies - II. Quantifying the effects of stochasticity on star formation rate indicators. , keywords =. doi:10.1093/mnras/stu1688 , archivePrefix =. 1403.4605 , primaryClass =

    Pith/arXiv arXiv

  12. [13]

    SLUG Stochastically Lighting Up Galaxies. I. Methods and Validating Tests. , keywords =. doi:10.1088/0004-637X/745/2/145 , archivePrefix =. 1106.3072 , primaryClass =

    Pith/arXiv arXiv

  13. [14]

    , keywords =

    Characterizing the Heavy Elements in Globular Cluster M22 and an Empirical s-process Abundance Distribution Derived from the Two Stellar Groups. , keywords =. doi:10.1088/0004-637X/742/1/37 , archivePrefix =. 1108.3868 , primaryClass =

    Pith/arXiv arXiv

  14. [15]

    , keywords =

    Proper Motions and Radial Velocities in the Globular Cluster M22 and the Cluster Distance. , keywords =. doi:10.1086/173590 , adsurl =

    work page doi:10.1086/173590

  15. [16]

    , keywords =

    CNO and F abundances in the globular cluster M 22 (NGC 6656). , keywords =. doi:10.1051/0004-6361/201118623 , archivePrefix =. 1202.0797 , primaryClass =

    Pith/arXiv arXiv

  16. [17]

    , keywords =

    Core-collapse Supernovae from 9 to 120 Solar Masses Based on Neutrino-powered Explosions. , keywords =. doi:10.3847/0004-637X/821/1/38 , archivePrefix =. 1510.04643 , primaryClass =

    Pith/arXiv arXiv

  17. [18]

    Nucleosynthesis and yields - Z = 0.02, 0.008 and 0.004

    Super and massive AGB stars - II. Nucleosynthesis and yields - Z = 0.02, 0.008 and 0.004. , keywords =. doi:10.1093/mnras/stt1877 , archivePrefix =. 1310.2614 , primaryClass =

    Pith/arXiv arXiv

  18. [19]

    , keywords =

    Stellar Yields from Metal-rich Asymptotic Giant Branch Models. , keywords =. doi:10.3847/0004-637X/825/1/26 , archivePrefix =. 1604.02178 , primaryClass =

    Pith/arXiv arXiv

  19. [20]

    , keywords =

    The impact of starbursts on element abundance ratios. , keywords =. doi:10.1093/mnras/staa2431 , archivePrefix =. 1911.02598 , primaryClass =

    Pith/arXiv arXiv 1911

  20. [21]

    arXiv e-prints , keywords =

    Stellar Migration and Chemical Enrichment in the Milky Way Disc: A Hybrid Model. arXiv e-prints , keywords =

  21. [22]

    , keywords =

    The Similarity of Abundance Ratio Trends and Nucleosynthetic Patterns in the Milky Way Disk and Bulge. , keywords =. doi:10.3847/1538-4357/abd6be , archivePrefix =. 2009.05063 , primaryClass =

    Pith/arXiv arXiv 2009

  22. [23]

    , keywords =

    Abundance Ratios in GALAH DR2 and Their Implications for Nucleosynthesis. , keywords =. doi:10.3847/1538-4357/ab4b5d , archivePrefix =. 1908.06113 , primaryClass =

    Pith/arXiv arXiv 1908

  23. [24]

    , keywords =

    The Origin of Elements from Carbon to Uranium. , keywords =. doi:10.3847/1538-4357/abae65 , archivePrefix =. 2008.04660 , primaryClass =

    Pith/arXiv arXiv 2008

  24. [25]

    arXiv e-prints , keywords =

    Origin of metals in old Milky Way halo stars based on GALAH and Gaia. arXiv e-prints , keywords =

  25. [26]

    , keywords =

    Chemical Cartography with APOGEE: Multi-element Abundance Ratios. , keywords =. doi:10.3847/1538-4357/ab07c7 , archivePrefix =. 1810.12325 , primaryClass =

    Pith/arXiv arXiv

  26. [27]

    arXiv e-prints , keywords =

    The GALAH Survey: Dependence of elemental abundances on age and metallicity for stars in the Galactic disc. arXiv e-prints , keywords =

  27. [28]

    , keywords =

    Simulating Metal Mixing of Both Common and Rare Enrichment Sources in a Low-mass Dwarf Galaxy. , keywords =. doi:10.3847/1538-4357/ab6efc , archivePrefix =. 1909.04695 , primaryClass =

    Pith/arXiv arXiv 1909

  28. [29]

    , keywords =

    Mixing of metals during star cluster formation: statistics and implications for chemical tagging. , keywords =. doi:10.1093/mnras/sty2625 , archivePrefix =. 1807.01712 , primaryClass =

    Pith/arXiv arXiv

  29. [30]

    Simulating the multiphase ISM of a dwarf galaxy with variable energy supernovae from individual stars

    LYRA - I. Simulating the multiphase ISM of a dwarf galaxy with variable energy supernovae from individual stars. , keywords =. doi:10.1093/mnras/staa3875 , archivePrefix =. 2010.07311 , primaryClass =

    Pith/arXiv arXiv 2010

  30. [31]

    , keywords =

    Dust-regulated galaxy formation and evolution: a new chemodynamical model with live dust particles. , keywords =. doi:10.1093/mnras/stv165 , archivePrefix =. 1501.05459 , primaryClass =

    Pith/arXiv arXiv

  31. [32]

    , keywords =

    The New Galaxy: Signatures of Its Formation. , keywords =. doi:10.1146/annurev.astro.40.060401.093840 , archivePrefix =. astro-ph/0208106 , primaryClass =

    arXiv

  32. [33]

    , keywords =

    Detection of metallicity correlations in 100 nearby galaxies. , keywords =. doi:10.1093/mnras/stab1263 , archivePrefix =. 2104.14807 , primaryClass =

    Pith/arXiv arXiv

  33. [34]

    , keywords =

    Measuring the mixing scale of the ISM within nearby spiral galaxies. , keywords =. doi:10.1093/mnras/staa2743 , archivePrefix =. 2009.02342 , primaryClass =

    Pith/arXiv arXiv 2009

  34. [35]

    , keywords =

    Mapping Metallicity Variations across Nearby Galaxy Disks. , keywords =. doi:10.3847/1538-4357/ab5115 , archivePrefix =. 1910.07190 , primaryClass =

    Pith/arXiv arXiv 1910

  35. [36]

    , keywords =

    3D gas-phase elemental abundances across the formation histories of Milky Way-mass galaxies in the FIRE simulations: initial conditions for chemical tagging. , keywords =. doi:10.1093/mnras/stab1606 , archivePrefix =. 2102.06220 , primaryClass =

    Pith/arXiv arXiv

  36. [37]

    , keywords =

    Numerical convergence of hydrodynamical simulations of galaxy formation: the abundance and internal structure of galaxies and their cold dark matter haloes. , keywords =. doi:10.1093/mnras/staa316 , archivePrefix =. 1908.05019 , primaryClass =

    Pith/arXiv arXiv 1908

  37. [38]

    The Solar Twin Planet Search. I. Fundamental parameters of the stellar sample. , keywords =. doi:10.1051/0004-6361/201424244 , archivePrefix =. 1408.4130 , primaryClass =

    Pith/arXiv arXiv

  38. [39]

    , year = 1973, month = sep, volume =

    The nitrogen abundance of the very metal-poor star HD 122563. , year = 1973, month = sep, volume =. doi:10.1086/152374 , adsurl =

    doi:10.1086/152374 1973

  39. [40]

    The chemical composition of red giants in 47 Tucanae. II. Magnesium isotopes and pollution scenarios. , keywords =. doi:10.1051/0004-6361/201526643 , archivePrefix =. 1602.00058 , primaryClass =

    Pith/arXiv arXiv

  40. [41]

    , keywords =

    Magnesium Isotope Ratios in Centauri Red Giants. , keywords =. doi:10.1088/0004-637X/769/1/8 , archivePrefix =. 1304.0523 , primaryClass =

    Pith/arXiv arXiv

  41. [42]

    , keywords =

    DAOSPEC: An Automatic Code for Measuring Equivalent Widths in High-Resolution Stellar Spectra. , keywords =. doi:10.1086/596126 , archivePrefix =. 0811.2932 , primaryClass =

    Pith/arXiv arXiv

  42. [43]

    , keywords =

    Magnesium Isotope Ratios in Hyades Stars. , keywords =. doi:10.1086/381701 , archivePrefix =. astro-ph/0312054 , primaryClass =

    Pith/arXiv arXiv

  43. [44]

    Abundances in globular cluster red giants. IV. M 22 and omega Centauri. , keywords =. doi:10.1086/159097 , adsurl =

    doi:10.1086/159097

  44. [45]

    , keywords =

    Gaia and Hubble Unveil the Kinematics of Stellar Populations in the Type II Globular Clusters Centauri and M22. , keywords =. doi:10.3847/1538-4357/aba04b , archivePrefix =. 2006.16355 , primaryClass =

    Pith/arXiv arXiv 2006

  45. [46]

    , keywords =

    M22: An [Fe/H] Abundance Range Revealed. , keywords =. doi:10.1088/0004-637X/705/2/1481 , archivePrefix =. 0909.5265 , primaryClass =

    Pith/arXiv arXiv

  46. [47]

    Star Clusters , year = 1980, editor =

    Evidence for primordial inhomogeneities from abundance of giants in M 5, M 13, and M 22. Star Clusters , year = 1980, editor =

    1980

  47. [48]

    , keywords =

    Further spectroscopic evidence bearing on the M22-omega Centauri comparison. , keywords =. doi:10.1086/157524 , adsurl =

    work page doi:10.1086/157524

  48. [49]

    , year = 1977, month = apr, volume =

    Cyanogen strengths and ultraviolet excesses of evolved stars in 17 globular clusters from DDO photometry. , year = 1977, month = apr, volume =. doi:10.1086/190438 , adsurl =

    work page doi:10.1086/190438 1977

  49. [50]

    , keywords =

    Cyanogen strengths of globular cluster post-main-sequence stars. , keywords =. doi:10.1086/182192 , adsurl =

    work page doi:10.1086/182192

  50. [51]

    , year = 1975, month = oct, volume =

    The Chemical Inhomogeneity of Omega Centauri. , year = 1975, month = oct, volume =. doi:10.1086/181945 , adsurl =

    doi:10.1086/181945 1975

  51. [52]

    , year = 1975, month = oct, volume =

    Abundance variations on the lower giant branch of Omega Centauri. , year = 1975, month = oct, volume =. doi:10.1086/181946 , adsurl =

    work page doi:10.1086/181946 1975

  52. [53]

    The anomalous giant branch of omega Centauri

    Near infrared photometry of globular clusters - VIII. The anomalous giant branch of omega Centauri. , keywords =. doi:10.1093/mnras/181.3.591 , adsurl =

    work page doi:10.1093/mnras/181.3.591

  53. [54]

    , keywords =

    Membership and other problems of the globular cluster M22. , keywords =. doi:10.1093/mnras/182.2.293 , adsurl =

    work page doi:10.1093/mnras/182.2.293

  54. [55]

    , year = 1976, month = oct, volume =

    Barium stars in the globular cluster NGC 6656. , year = 1976, month = oct, volume =. doi:10.1093/mnras/177.1.73 , adsurl =

    work page doi:10.1093/mnras/177.1.73 1976

  55. [56]

    , year = 1976, month = apr, volume =

    Evidence for Abundance Differences amongh Giant Branch Stars in the Globular Cluster Omega Centauri. , year = 1976, month = apr, volume =

    1976

  56. [57]

    The Observatory , year = 1962, month = oct, volume =

    A CH star in omega Centauri. The Observatory , year = 1962, month = oct, volume =

    1962

  57. [58]

    , keywords =

    A new CH star in the globular cluster Omega Centauri. , keywords =. doi:10.1086/181977 , adsurl =

    work page doi:10.1086/181977

  58. [59]

    Royal Observatory Annals , keywords =

    Studies of the globular cluster [omega] Centauri I,. Royal Observatory Annals , keywords =

  59. [60]

    Photometry of Southern globular clusters. I. Bright stars in OME Cen. , year = 1973, month = jan, volume =. doi:10.1093/mnras/162.3.207 , adsurl =

    work page doi:10.1093/mnras/162.3.207 1973

  60. [61]

    The Giant Branch of Omega Centauri. I. Abundance Variations due to Mixing. , year = 1976, month = sep, volume =. doi:10.1086/154616 , adsurl =

    work page doi:10.1086/154616 1976

  61. [62]

    , keywords =

    Wide-field photometry of the Galactic globular cluster M22. , keywords =. doi:10.1111/j.1365-2966.2004.07599.x , archivePrefix =. astro-ph/0401392 , primaryClass =

    arXiv 2004

  62. [63]

    , keywords =

    Faint nebulosity near OME Cen. , keywords =

  63. [64]

    , keywords =

    Str \"o mgren photometry in globular clusters: M55 & M22. , keywords =

  64. [65]

    , keywords =

    The chemical inhomogeneity of M 22. , keywords =. doi:10.1086/160764 , adsurl =

    work page doi:10.1086/160764

  65. [66]

    Vistas in Astronomy , year = 1956, month = jan, volume =

    The dimensions of omega centauri. Vistas in Astronomy , year = 1956, month = jan, volume =. doi:10.1016/0083-6656(56)90032-0 , adsurl =

    work page doi:10.1016/0083-6656(56)90032-0 1956

  66. [67]

    , year = 1961, month = mar, volume =

    The shape of a rotating star cluster. , year = 1961, month = mar, volume =. doi:10.1086/108376 , adsurl =

    work page doi:10.1086/108376 1961

  67. [68]

    , keywords =

    Rotating Globular Clusters. , keywords =. doi:10.1088/0004-637X/772/1/67 , archivePrefix =. 1305.6025 , primaryClass =

    Pith/arXiv arXiv

  68. [69]

    , keywords =

    The internal rotation of globular clusters revealed by Gaia DR2. , keywords =. doi:10.1093/mnras/sty2365 , archivePrefix =. 1806.02580 , primaryClass =

    Pith/arXiv arXiv

  69. [70]

    , keywords =

    A stellar census in globular clusters with MUSE: The contribution of rotation to cluster dynamics studied with 200 000 stars. , keywords =. doi:10.1093/mnras/stx2719 , archivePrefix =. 1710.07257 , primaryClass =

    Pith/arXiv arXiv

  70. [71]

    , keywords =

    Three-component Kinematics of Multiple Stellar Populations in Globular Clusters with Gaia and VLT. , keywords =. doi:10.3847/1538-4357/ab5aee , archivePrefix =. 1905.09908 , primaryClass =

    Pith/arXiv arXiv 1905

  71. [72]

    , keywords =

    Iron and s-elements abundance variations in NGC 5286: comparison with `anomalous' globular clusters and Milky Way satellites. , keywords =. doi:10.1093/mnras/stv420 , archivePrefix =. 1502.07438 , primaryClass =

    Pith/arXiv arXiv

  72. [73]

    , keywords =

    An Abundance Analysis of Bright Giants in the Globular Cluster NGC 1851. , keywords =. doi:10.1086/525850 , archivePrefix =. 0711.1394 , primaryClass =

    Pith/arXiv arXiv

  73. [74]

    , keywords =

    Iron and neutron-capture element abundance variations in the globular cluster M2 (NGC 7089) ^ ★. , keywords =. doi:10.1093/mnras/stu806 , archivePrefix =. 1404.6873 , primaryClass =

    Pith/arXiv arXiv

  74. [75]

    , keywords =

    Formation of Centauri from an ancient nucleated dwarf galaxy in the young Galactic disc. , keywords =. doi:10.1046/j.1365-2966.2003.07275.x , archivePrefix =. astro-ph/0310348 , primaryClass =

    arXiv 2003

  75. [76]

    , keywords =

    A catalogue of masses, structural parameters, and velocity dispersion profiles of 112 Milky Way globular clusters. , keywords =. doi:10.1093/mnras/sty1057 , archivePrefix =. 1804.08359 , primaryClass =

    Pith/arXiv arXiv

  76. [77]

    , keywords =

    The Gaia-ESO Survey: Kinematics of seven Galactic globular clusters. , keywords =. doi:10.1051/0004-6361/201425036 , archivePrefix =. 1411.4886 , primaryClass =

    Pith/arXiv arXiv

  77. [78]

    Catalogue des N \'e buleuses et des Amas d' \'E toiles (Catalog of Nebulae and Star Clusters)

  78. [79]

    The Globular Cluster Messier 22 (N. G. C. 6656). Publications of the American Astronomical Society , year = 1922, series =

    1922

  79. [80]

    , year = 1959, month = feb, volume =

    Color-magnitude diagram for the globular cluster M 22. , year = 1959, month = feb, volume =. doi:10.1086/107848 , adsurl =

    work page doi:10.1086/107848 1959

  80. [81]

    Harvard College Observatory Bulletin , year = 1930, month = mar, volume =

    The Mass-Spectrum Relation for Giant Stars in the Globular Cluster Messier 22. Harvard College Observatory Bulletin , year = 1930, month = mar, volume =

    1930

Showing first 80 references.