Pith. sign in

REVIEW 3 major objections 4 minor 100 references

Standard He II diagnostics find pure Pop III galaxies but miss hybrids once Pop II stars contribute much of the mass.

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 · grok-4.5

2026-07-10 17:06 UTC pith:44QG2HXE

load-bearing objection Clean proof-of-concept showing pure-Pop-III HeII cuts fail for hybrids; the TNG re-labeling is the known soft spot, not a hidden flaw. the 3 major comments →

arxiv 2607.07832 v1 pith:44QG2HXE submitted 2026-07-08 astro-ph.GA

On the detection of Population III galaxies: Emission Line Diagnostics for Hybrid Stellar Populations

S. Goswami , J. M. V\'ilchez , C. Kehrig , A. Ferrara , J. Iglesias-P\'aramo This is my paper
classification astro-ph.GA
keywords Population III starsemission-line diagnosticshybrid stellar populationsHe IIphotoionization modelsreionizationIllustrisTNGhigh-redshift galaxies
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 first metal-free stars (Population III) are expected to leave hard-ionization fingerprints such as strong He II emission, yet real galaxies can contain both those stars and later metal-enriched (Population II) stars. This paper takes galaxies drawn from a large cosmological simulation, re-labels the most metal-poor stellar particles as Pop III, and computes their integrated spectra with photoionization models that use the local gas density and luminosity of each simulation cell. The resulting line ratios and equivalent widths show that only systems almost entirely made of Pop III stars land cleanly inside the regions previously proposed as Pop III diagnostics. Once Pop II stars supply a substantial fraction of the mass, metal lines appear, He II ratios drop, and the same objects fall into ambiguous or Pop II-like territory. The work therefore argues that existing emission-line criteria are incomplete for hybrid populations and that new diagnostics sensitive to mixed stellar content are required if residual Pop III stars are to be identified during and after reionization.

Core claim

Emission-line diagnostics that rely on He II strength relative to hydrogen or continuum are strongly dependent on the Pop III-to-total stellar mass ratio: they cleanly select pure Pop III systems but introduce degeneracies and fail to flag galaxies once Pop II stars contribute significantly, so hybrid populations can be misclassified or missed.

What carries the argument

Self-consistent post-processing of simulated galaxies: extremely metal-poor star particles are assigned Pop III spectral energy distributions, Cloudy photoionization models are run on each host cell using its local density and luminosity, and the cell spectra are stacked to produce galaxy-integrated emission-line ratios that are then placed on published diagnostic diagrams as a function of Pop III mass fraction.

Load-bearing premise

The simulation itself never forms true Population III stars, so the authors must re-label the most metal-poor particles as Pop III proxies and graft on external spectral models; if those proxies do not match the real ages, locations or ionizing output of residual Pop III stars, the hybrid spectra are not representative.

What would settle it

Obtain high-signal spectra of a statistically useful sample of high-redshift or local extreme-emission galaxies that independently show both strong He II and clear metal lines, measure their Pop III mass fractions by other means (for example deep rest-UV continuum or abundance patterns), and check whether those hybrids systematically fall outside the pure-Pop-III diagnostic boxes exactly as the models predict.

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

If this is right

  • Pure Pop III candidates will continue to be recoverable with existing He II equivalent-width and line-ratio cuts, but hybrid systems will be under-counted.
  • Complementary ratios such as [Ne V]/[Ne III] and He I 5876/Heta can help flag mixed populations that standard He II diagnostics miss.
  • Diagnostic templates used for JWST and ELT surveys of reionization-era galaxies need to include a continuous range of Pop III mass fractions rather than pure end-members.
  • The same hybrid ambiguity can affect interpretation of local metal-poor dwarfs that already show puzzling He II emission.

Where Pith is reading between the lines

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

  • If inefficient metal mixing really allows pristine pockets to persist to low redshift, a non-negligible fraction of He II emitters already catalogued may be hybrids rather than pure Pop III or pure AGN, altering demographic estimates of the first stars.
  • Future large-volume simulations that form Pop III self-consistently could re-run the identical diagnostic exercise and immediately test whether the mass-fraction dependence survives when the stars and gas evolve together.
  • The geometry and IMF sensitivity shown here implies that plane-parallel versus spherical assumptions and top-heavy versus Kroupa-like IMFs must be marginalized when claiming a Pop III detection from a single spectrum.

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

3 major / 4 minor

Summary. The paper is a proof-of-concept study of how hybrid Pop III/Pop II stellar populations affect standard emission-line diagnostics used to identify metal-free stars. Galaxies with Pop III-to-total mass ratios of 1.0, 0.8, 0.5 and 0.1 are selected from IllustrisTNG50-1 by re-labelling extremely metal-poor star particles (Z_init < 4e-4 Z_sun) as Pop III proxies. Yggdrasil SEDs (three Pop III IMFs) are assigned in post-processing and Cloudy photoionization models are run cell-by-cell using local n_H and cell sizes from the simulation; the resulting spectra are stacked. The authors show that pure-Pop-III systems fall inside the Nakajima & Maiolino (2022) EW(He II)–line-ratio loci, while systems with substantial Pop II contributions fall outside them, and they explore complementary diagnostics (He II/Heta vs [Ne V]/[Ne III], EW(He II 1640) vs He I 5876/Heta). IMF and geometry variations are tested and the limitations of the TNG50 proxy approach are stated explicitly.

Significance. If the qualitative result holds, it is a useful caution for JWST-era searches for Pop III signatures: existing pure-Pop-III cuts can miss or misclassify hybrid systems that cosmological simulations suggest may be common. The work is transparent about its post-processing nature, explores the main free parameters (IMF, geometry, age cut), and places its models in the same diagnostic space as recent JWST candidates (e.g. Hebe). Strengths include the self-consistent use of local cell densities and sizes, the explicit comparison of three Pop III IMFs, and the open framing as a proof-of-concept rather than a predictive census. The five-galaxy sample is deliberately extreme, so the paper’s value is illustrative rather than statistical.

major comments (3)
  1. The central claim rests on only five deliberately selected youngest systems (Table 1). While the qualitative placement of pure versus hybrid systems is clear in Figs. 5–6 and 8, the manuscript does not quantify how common such hybrid configurations are even within TNG50, nor does it show that the same diagnostic failure occurs for a broader, less age-selected sample. A short additional test (or explicit statement that the result is intended only as an existence proof) would strengthen the load-bearing claim that “current diagnostics fail for systems where Pop II stars contribute significantly.”
  2. Sect. 2 and 4.3 correctly note that TNG50 has no Pop III formation or feedback physics, so the metal-poor particles are only proxies. The ages, spatial clustering and local n_H of these particles therefore reflect ordinary low-Z star formation rather than true Pop III minihalo conditions. Because the He II strength is highly age-sensitive (Fig. 1), any systematic difference in the age distribution of real residual Pop III stars would move the hybrid points. The paper already labels itself a proof-of-concept; it should state more explicitly that the quantitative line ratios are not predictions for real Pop III hybrids and that the main result is the qualitative degeneracy once Pop II contributes.
  3. Pop II particles are assigned a single metallicity Z = 0.02 (Sect. 2.2). The authors argue this maximises the Pop II He II contribution, but the resulting metal-line strengths (and therefore the He II / metal-line ratios in Fig. 4) are upper limits. A brief check at a lower Pop II metallicity (or a short statement of how the diagnostic placements would shift) would confirm that the hybrid systems still leave the pure-Pop-III loci.
minor comments (4)
  1. Fig. 3 caption and the SG5 panel label both say “z = 6.0” while Table 1 and the text correctly give z = 0 for SG5; this should be corrected.
  2. The definition of Z_crit = 0.02 Z_⊙^{2} = 4 × 10^{-4} Z_⊙ is written inconsistently in places (sometimes as 0.02 Z_⊙^{2}, sometimes as 0.02 Z_⊙); a single clear expression would help.
  3. In Sect. 4 the age-cut test is mentioned only briefly; quoting the maximum change in He II EW and He II/Hβ (already given as ~0.04–0.9 dex) in a short table or sentence would make the robustness check easier to evaluate.
  4. A few references appear with future arXiv numbers (e.g. 2026 papers); ensure the final citation list is consistent with the journal’s style once those works are published or updated.

Circularity Check

0 steps flagged

No significant circularity: hybrid diagnostic placements follow from external TNG selection + Yggdrasil/Cloudy post-processing against independent literature cuts.

full rationale

The paper's central claim (pure-Pop-III cuts succeed only for M_PopIII/M_tot=1 systems and fail once Pop II contributes substantially) is obtained by (i) selecting TNG50 galaxies via an external metallicity threshold Z_crit=4e-4 Z_sun taken from Zackrisson/Bromm literature, (ii) assigning pre-computed Yggdrasil SEDs for three fixed IMFs, (iii) running Cloudy on local cell n_H and L_tot, and (iv) placing the resulting line ratios/EWs on diagnostic diagrams taken verbatim from Nakajima & Maiolino (2022), Chisholm et al. (2024) and Shirazi & Brinchmann (2012). No free parameters are fitted to force galaxies into or out of the shaded regions; IMF and geometry variations are explored and leave the qualitative conclusion unchanged. Self-citations (local He II work by Kehrig/Vílchez, comparison papers by Venditti/Rusta) are used only for context or external comparison and do not underwrite the derivation. The acknowledged proxy nature of TNG Pop-III-like particles is a modeling limitation, not a circular reduction of output to input. The derivation chain is therefore self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

4 free parameters · 3 axioms · 0 invented entities

The central claim rests on a post-processing pipeline that reinterprets TNG star particles as Pop III or Pop II according to a metallicity threshold, assigns external SEDs, and runs Cloudy under idealized abundance and geometry assumptions. These choices are free parameters or domain assumptions rather than derived results; the paper itself flags most of them in Sect. 4.3.

free parameters (4)
  • Z_crit = 4e-4 Z_sun
    Critical metallicity separating Pop III from Pop II particles set by hand to 0.02 Z_sun^2 = 4e-4 Z_sun (Sect. 2.1). Different literature values shift the hybrid fractions and therefore the diagnostic locations.
  • Pop III IMF choice = PopIII.1 (fiducial)
    Three discrete IMFs (PopIII.1 Salpeter 50-500, PopIII.2 log-normal, PopIII.K Kroupa) are explored; the most top-heavy is used as fiducial. Absolute HeII strengths scale strongly with this choice (Fig. 7).
  • Pop II metallicity = 0.02
    All Pop II particles assigned a single Z=0.02 SED regardless of the particle's actual metallicity in TNG (Sect. 2.2, 4.3). Chosen as a conservative upper bound on WR contribution.
  • nebular radius = R_cell/2
    Cloudy integration stopped at R/2 where R is the TNG cell size; geometry (spherical vs plane-parallel) is also free (Sect. 2.2).
axioms (3)
  • ad hoc to paper Extremely metal-poor star particles in TNG50 can be treated as Pop III proxies even though the simulation has no Pop III formation or feedback physics.
    Stated explicitly in Sect. 2 and 4.3; the entire hybrid sample rests on this re-labeling.
  • domain assumption Nebular emission is ionization-bounded and can be computed cell-by-cell with constant density, no dust for Pop III, and primordial abundances only.
    Standard Cloudy setup adopted in Sect. 2.2; geometry and density variations are tested but the constant-density assumption is not.
  • domain assumption Youngest galaxies at each redshift dominate the HeII signal; older particles contribute negligibly.
    Selection criterion in Sect. 2.1 and Fig. 1; age-cut tests later confirm only modest quantitative shifts.

pith-pipeline@v1.1.0-grok45 · 30609 in / 3151 out tokens · 37605 ms · 2026-07-10T17:06:31.110184+00:00 · methodology

0 comments
read the original abstract

Identifying Population III (Pop III) stars, the first generation of metal-free stars in the early Universe, remains a central challenge in astrophysics. High-ionization emission lines, such as HeII are commonly used as tracers of Pop III signatures. However, realistic galaxies may host hybrid stellar populations, including both Pop III and metal-enriched Population II (Pop II) stars, complicating the interpretation of observed spectra. We aim to investigate how hybrid Pop III/Pop II stellar populations affect emission line diagnostics and assess the detectability of Pop III stars across different galactic environments and redshifts. We select galaxies with varying Pop III-to-total mass ratios from the IllustrisTNG cosmological simulations. Using self-consistent photoionisation models, we compute integrated spectra by adopting local physical conditions from the simulations to study the resulting emission line diagnostics. We find that emission line diagnostics are strongly dependent on the relative Pop III contribution. Current diagnostics can identify galaxies dominated totally by Pop III stars but fail for systems where Pop II stars contribute significantly, introducing degeneracies in interpreting observed spectra. Our results highlight the limitations of existing emission line diagnostics in hybrid systems and emphasize the need for additional methods that account for mixed stellar populations to reliably detect Pop III stars during and after the epoch of reionization.

Figures

Figures reproduced from arXiv: 2607.07832 by A. Ferrara, C. Kehrig, J. Iglesias-P\'aramo, J. M. V\'ilchez, S. Goswami.

Figure 1
Figure 1. Figure 1: Evolution of the EW of the He II emission lines as a function of stellar population age for Population III models. The left panel shows the EW of the nebular He II𝜆4686 line, while the right panel presents the EW of the HeII𝜆1640 line. times using the astropy.cosmology framework. This method ensures that our sample preferentially includes systems whose stellar populations are dominated by recently formed v… view at source ↗
Figure 2
Figure 2. Figure 2: Gas column density maps of the five galaxies selected, SG1, SG2, SG3, SG4, and SG5 (from left to right, respectively), from the TNG-50-1 simulations. The gas column density values are displayed in the color bar on the right, while the physical scales are displayed at the upper left of each galaxy. 1000 2000 3000 4000 5000 6000 7000 Wavelength (Å) 10 1 10 0 10 1 10 2 Flux (normalized) H 3968 H (4340 ) H (48… view at source ↗
Figure 3
Figure 3. Figure 3: Integrated spectrum of the five simulated galaxies in this work post-Cloudy asssuming a PopIII.1 IMF and a spherical geometry. All the spectra are normalized to the flux at 1500 Å, Flux1500 . Some key emission lines to distinguish between PopIII and PopII characteristics are marked in red shaded regions. Every star particle in TNG-50-1 has an assigned initial mass, 𝑀★,initial. However, TNG50 does not inclu… view at source ↗
Figure 4
Figure 4. Figure 4: Flux ratios of the He ii𝜆1640 line with respect to Flux1500, [OIII]𝜆1661 and CIV𝜆1549 for the five galaxies studied in this work. Galaxies SG2, SG3 and SG4 are at 𝑧=6 while SG1 is at 𝑧=10 and SG5 is at 𝑧=0. Since 100 % Pop III systems like SG1 and SG2 don’t produce metal lines such as [OIII]𝜆1661 and CIV𝜆1549, ratios involving these lines are not shown for these galaxies. The ratio of 𝑀∗,PopIII/𝑀∗,total, w… view at source ↗
Figure 6
Figure 6. Figure 6: Left: The He ii𝜆4686/H𝛽 vs. [Ne V]𝜆3427/[Ne III]𝜆3869 emis￾sion line diagnostic diagram. The grey horizontal line shown is proposed by Shirazi & Brinchmann (2012) and the vertical line instead is from Cleri et al. (2023) to separate AGN and SF galaxies. The cyan triangles are a local sample of SF galaxies taken from Izotov et al. (2021). GN 42 437 is denoted by the brown diamond and taken from Chisholm et … view at source ↗
Figure 7
Figure 7. Figure 7: Effect of different IMFs on the flux ratios for our simulated galaxies. Each panel corresponds to a different galaxy; symbols are the same as in [PITH_FULL_IMAGE:figures/full_fig_p008_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Left: EW(He ii 𝜆4686) vs. He ii/H𝛽 for different systems. Right: Same as the left panel, but for EW(He ii 𝜆1640) vs. He ii/Ly𝛼. The top panels are related to models using the Pop III.2 IMF whereas the bottom panels are for the case of Pop III.K IMF. The symbols shown are the same as in [PITH_FULL_IMAGE:figures/full_fig_p009_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Same as [PITH_FULL_IMAGE:figures/full_fig_p009_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: EW of He ii 𝜆1640 as a function of the He ii 𝜆1640/H𝛾 line ratio for different ionizing sources. The JWST Pop III candidate Hebe (total) and its two components are shown in yellow stars, whereas photoioniza￾tion models from Nakajima & Maiolino (2022) are shown in blue stars (Pop III), red crosses (Pop II), and green diamonds (AGN). The five simulated galaxies used in this work are shown as red circles. Po… 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

100 extracted references · 100 canonical work pages · 7 internal anchors

  1. [1]

    M., et al

    Arroyo-Polonio, A., Kehrig, C., Vílchez, J. M., et al. 2025, ApJ, 987, L36

    work page 2025

  2. [2]

    A., Chisholm, J., Erb, D

    Berg, D. A., Chisholm, J., Erb, D. K., et al. 2019, ApJ, 878, L3

    work page 2019

  3. [3]

    A., Chisholm, J., Erb, D

    Berg, D. A., Chisholm, J., Erb, D. K., et al. 2021, ApJ, 922, 170

    work page 2021

  4. [4]

    2021, ApJ, 918, 5

    Berzin, E., Secunda, A., Cen, R., Menegas, A., & Götberg, Y. 2021, ApJ, 918, 5

    work page 2021

  5. [5]

    C., Stanway, E

    Bray, J. C., Stanway, E. R., & Eldridge, J. J. 2025, MNRAS, 542, 2087

    work page 2025

  6. [6]

    S., & Larson, R

    Bromm, V., Coppi, P. S., & Larson, R. B. 2002, ApJ, 564, 23

    work page 2002

  7. [7]

    S., & Larson, R

    Bromm, V., Ferrara, A., Coppi, P. S., & Larson, R. B. 2001, MNRAS, 328, 969

    work page 2001

  8. [8]

    & Larson, R

    Bromm, V. & Larson, R. B. 2004, ARA&A, 42, 79

    work page 2004

  9. [9]

    & Loeb, A

    Bromm, V. & Loeb, A. 2003, Nature, 425, 812

    work page 2003

  10. [10]

    2025, ApJ, 993, L52

    Cai, S., Li, M., Cai, Z., et al. 2025, ApJ, 993, L52

    work page 2025

  11. [11]

    Ceverino, D., Glover, S. C. O., & Klessen, R. S. 2017, MNRAS, 470, 2791

    work page 2017

  12. [12]

    2003, PASP, 115, 763

    Chabrier, G. 2003, PASP, 115, 763

    work page 2003

  13. [13]

    A., Endsley, R., et al

    Chisholm, J., Berg, D. A., Endsley, R., et al. 2024, MNRAS, 534, 2633

    work page 2024

  14. [14]

    & Ferrara, A

    Ciardi, B. & Ferrara, A. 2005, Space Sci. Rev., 116, 625

    work page 2005

  15. [15]

    J., Olivier, G

    Cleri, N. J., Olivier, G. M., Hutchison, T. A., et al. 2023, ApJ, 953, 10

    work page 2023

  16. [16]

    Crowther, P. A. 2007, ARA&A, 45, 177

    work page 2007

  17. [17]

    C., Scholte, D., et al

    Cullen, F., Carnall, A. C., Scholte, D., et al. 2025, MNRAS, 540, 2176 Curtis-Lake,E.,Carniani,S.,Cameron,A.,etal.2023,NatureAstronomy,7,622

    work page 2025

  18. [18]

    S., & White, S

    Davis, M., Efstathiou, G., Frenk, C. S., & White, S. D. M. 1985, ApJ, 292, 371

    work page 1985

  19. [19]

    J., Izzard, R

    Eldridge, J. J., Izzard, R. G., & Tout, C. A. 2008, MNRAS, 384, 1109

    work page 2008

  20. [20]

    Eldridge, J. J. & Stanway, E. R. 2022, ARA&A, 60, 455

    work page 2022

  21. [21]

    2016, MNRAS, 456, 3354

    Feltre, A., Charlot, S., & Gutkin, J. 2016, MNRAS, 456, 3354

    work page 2016

  22. [22]

    J., Chatzikos, M., Guzmán, F., et al

    Ferland, G. J., Chatzikos, M., Guzmán, F., et al. 2017, Rev. Mexicana Astron. Astrofis., 53, 385

    work page 2017

  23. [23]

    J., Porter, R

    Ferland, G. J., Porter, R. L., van Hoof, P. A. M., et al. 2013, Rev. Mexicana Astron. Astrofis., 49, 137

    work page 2013

  24. [24]

    2013, ApJ, 764, 41

    Fragos, T., Lehmer, B., Tremmel, M., et al. 2013, ApJ, 764, 41

    work page 2013

  25. [25]

    P., Chisholm, J., et al

    Fujimoto, S., Naidu, R. P., Chisholm, J., et al. 2025, ApJ, 989, 46

    work page 2025

  26. [26]

    R., Kennicutt, Jr., R

    Garnett, D. R., Kennicutt, Jr., R. C., Chu, Y.-H., & Skillman, E. D. 1991, ApJ, 373, 458 González-Díaz, R., Vílchez, J. M., Kehrig, C., del Moral-Castro, I., & Iglesias- Páramo, J. 2025, arXiv e-prints, arXiv:2506.11685

    work page arXiv 1991

  27. [27]

    2020, MNRAS, 494, 1071

    Graziani, L., Schneider, R., Ginolfi, M., et al. 2020, MNRAS, 494, 1071

    work page 2020

  28. [28]

    F., Dooley, G

    Griffen, B. F., Dooley, G. A., Ji, A. P., et al. 2016, MNRAS, 456, 946

    work page 2016

  29. [29]

    2021, MNRAS, 501, 5517

    Grisdale, K., Thatte, N., Devriendt, J., et al. 2021, MNRAS, 501, 5517

    work page 2021

  30. [30]

    2016, MNRAS, 462, 1757

    Gutkin, J., Charlot, S., & Bruzual, G. 2016, MNRAS, 462, 1757

    work page 2016

  31. [31]

    pySTARBURST99: The Next Generation of STARBURST99

    Hawcroft, C., Leitherer, C., Arangure, O., et al. 2025, arXiv e-prints, arXiv:2505.24841

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  32. [32]

    2014, ApJ, 781, 60

    Hirano, S., Hosokawa, T., Yoshida, N., et al. 2014, ApJ, 781, 60

    work page 2014

  33. [33]

    2017, MNRAS, 472, 2468

    Hirschmann, M., Charlot, S., Feltre, A., et al. 2017, MNRAS, 472, 2468

    work page 2017

  34. [34]

    & Kobayashi, C

    Hyung, M. & Kobayashi, C. 2017, MNRAS Letters, 467, L51

    work page 2017

  35. [35]

    Inoue, A. K. 2011, MNRAS, 415, 2920

    work page 2011

  36. [36]

    I., Thuan, T

    Izotov, Y. I., Thuan, T. X., & Guseva, N. G. 2021, MNRAS, 508, 2556

    work page 2021

  37. [37]

    L., & Bromm, V

    Jaacks, J., Thompson, R., Finkelstein, S. L., & Bromm, V. 2018, MNRAS, 475, 4396

    work page 2018

  38. [38]

    H., & Milosavljević, M

    Jeon, M., Bromm, V., Pawlik, A. H., & Milosavljević, M. 2015, MNRAS, 452, 1152

    work page 2015

  39. [39]

    B., Venditti, A., Bromm, V., et al

    Jeong, T. B., Venditti, A., Bromm, V., et al. 2026, arXiv e-prints, arXiv:2603.23209

    work page internal anchor Pith review arXiv 2026

  40. [40]

    & Haiman, Z

    Jimenez, R. & Haiman, Z. 2006, Nature, 440, 501

    work page 2006

  41. [41]

    L., Dalla Vecchia, C., & Khochfar, S

    Johnson, J. L., Dalla Vecchia, C., & Khochfar, S. 2013, MNRAS, 428, 1857

    work page 2013

  42. [42]

    2022, MNRAS, 511, 4005 Kehrig,C.,Guerrero,M.A.,Vílchez,J.M.,&Ramos-Larios,G.2021,ApJ,908, L54

    Kannan, R., Garaldi, E., Smith, A., et al. 2022, MNRAS, 511, 4005 Kehrig,C.,Guerrero,M.A.,Vílchez,J.M.,&Ramos-Larios,G.2021,ApJ,908, L54

    work page 2022

  43. [43]

    M., Guerrero, M

    Kehrig, C., Vílchez, J. M., Guerrero, M. A., et al. 2018, MNRAS, 480, 1081

    work page 2018

  44. [44]

    M., Pérez-Montero, E., et al

    Kehrig, C., Vílchez, J. M., Pérez-Montero, E., et al. 2015, ApJ, 801, L28

    work page 2015

  45. [45]

    J., Dopita, M

    Kewley, L. J., Dopita, M. A., Leitherer, C., et al. 2013, ApJ, 774, 100

    work page 2013

  46. [46]

    2023, MNRAS, 525, 190

    Koutsouridou, I., Salvadori, S., Skúladóttir, Á., et al. 2023, MNRAS, 525, 190

    work page 2023

  47. [47]

    2001, MNRAS, 322, 231

    Kroupa, P. 2001, MNRAS, 322, 231

    work page 2001

  48. [48]

    2025, A&A, 695, A17

    Lecroq, M., Charlot, S., Bressan, A., et al. 2025, A&A, 695, A17

    work page 2025

  49. [49]

    & Bromm, V

    Liu, B. & Bromm, V. 2020, MNRAS, 497, 2839

    work page 2020

  50. [50]

    F., Wetzel, A

    Ma, X., Hopkins, P. F., Wetzel, A. R., et al. 2017, MNRAS, 467, 2430

    work page 2017

  51. [51]

    2010, MNRAS, 407, 1003

    Maio, U., Ciardi, B., Dolag, K., Tornatore, L., & Khochfar, S. 2010, MNRAS, 407, 1003

    work page 2010

  52. [52]

    The search for Population III: Confirmation of a HeII emitter with no metal lines at z=10.6

    Maio, U., Petkova, M., De Lucia, G., & Borgani, S. 2016, MNRAS, 460, 3733 Maiolino,R.,Übler,H.,Perna,M.,etal.2026,arXive-prints,arXiv:2603.20362 Milosavljević, M., Bromm, V., Couch, S. M., & Oh, S. P. 2015, MNRAS, 452, 1152 Mingozzi,M.,GarciadelValle-Espinosa,M.,James,B.L.,etal.2025,ApJ,985, 253

    work page internal anchor Pith review Pith/arXiv arXiv 2016

  53. [53]

    Pristine Massive Star Formation Caught at the Break of Cosmic Dawn

    Mondal, C., Saha, K., Borgohain, A., et al. 2025, ApJ, 988, 171 Morishita,T.,Liu,Z.,Stiavelli,M.,etal.2025,arXive-prints,arXiv:2507.10521

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  54. [54]

    L., Gnedin, O

    Muratov, A. L., Gnedin, O. Y., Gnedin, N. Y., & Zemp, M. 2013, ApJ, 772, 106

    work page 2013

  55. [55]

    & Maiolino, R

    Nakajima, K. & Maiolino, R. 2022, MNRAS, 513, 5134

    work page 2022

  56. [56]

    & Ouchi, M

    Nakajima, K. & Ouchi, M. 2014, MNRAS, 442, 900

    work page 2014

  57. [57]

    2025, arXiv e-prints, arXiv:2506.11846

    Nakajima, K., Ouchi, M., Harikane, Y., et al. 2025, arXiv e-prints, arXiv:2506.11846

    work page arXiv 2025

  58. [58]

    2005, ApJ, 626, 627

    Omukai, K., Tsuribe, T., Schneider, R., & Ferrara, A. 2005, ApJ, 626, 627

    work page 2005

  59. [59]

    2014, MNRAS, 440, 2498 Pérez-Montero, E., Kehrig, C., Vílchez, J

    Pallottini, A., Ferrara, A., Gallerani, S., Salvadori, S., & D’Odorico, V. 2014, MNRAS, 440, 2498 Pérez-Montero, E., Kehrig, C., Vílchez, J. M., et al. 2020, A&A, 643, A80

    work page 2014

  60. [60]

    2019, MNRAS, 490, 3196

    Pillepich, A., Nelson, D., Springel, V., et al. 2019, MNRAS, 490, 3196

    work page 2019

  61. [61]

    2018, MNRAS, 473, 4077

    Pillepich, A., Springel, V., Nelson, D., et al. 2018, MNRAS, 473, 4077

    work page 2018

  62. [62]

    2019, MNRAS, 490, 978

    Plat, A., Charlot, S., Bruzual, G., et al. 2019, MNRAS, 490, 978

    work page 2019

  63. [63]

    Raiter, A., Schaerer, D., & Fosbury, R. A. E. 2010, A&A, 523, A64

    work page 2010

  64. [64]

    2024, ApJ, 976, 193

    Roberts-Borsani, G., Treu, T., Shapley, A., et al. 2024, ApJ, 976, 193

    work page 2024

  65. [65]

    2013, MNRAS, 436, 2188

    Rosdahl, J., Blaizot, J., Aubert, D., Stranex, T., & Teyssier, R. 2013, MNRAS, 436, 2188

    work page 2013

  66. [66]

    The Pristine HeII Emitter near GN-z11: Constraining the Mass Distribution of the First Stars

    Rusta, E., Salvadori, S., Maiolino, R., et al. 2026, arXiv e-prints, arXiv:2603.20363

    work page internal anchor Pith review Pith/arXiv arXiv 2026

  67. [67]

    Salpeter, E. E. 1955, ApJ, 121, 161

    work page 1955

  68. [68]

    2019, ApJ, 871, 206

    Sarmento, R., Scannapieco, E., & Côté, B. 2019, ApJ, 871, 206

    work page 2019

  69. [69]

    S., Fialkov, A., Hartwig, T., et al

    Sartorio, N. S., Fialkov, A., Hartwig, T., et al. 2023, MNRAS, 521, 4039

    work page 2023

  70. [70]

    2020, MNRAS, 496, 3796

    Saxena, A., Pentericci, L., Schaerer, D., et al. 2020, MNRAS, 496, 3796

    work page 2020

  71. [71]

    2002, A&A, 382, 28

    Schaerer, D. 2002, A&A, 382, 28

    work page 2002

  72. [72]

    2003, A&A, 397, 527

    Schaerer, D. 2003, A&A, 397, 527

    work page 2003

  73. [73]

    Schaerer, D., Fragos, T., & Izotov, Y. I. 2019, A&A, 622, L10

    work page 2019

  74. [74]

    & Stasińska, G

    Schaerer, D. & Stasińska, G. 1999, A&A, 345, L17 Schneider,R.,Omukai,K.,Bianchi,S.,&Valiante,R.2012,MNRAS,419,1566 Schneider,R.,Omukai,K.,Inoue,A.K.,&Ferrara,A.2006,MNRAS,369,1437

    work page 1999

  75. [75]

    P., Mirocha, J., et al

    Senchyna, P., Stark, D. P., Mirocha, J., et al. 2020, MNRAS, 494, 941

    work page 2020

  76. [76]

    P., Vidal-García, A., et al

    Senchyna, P., Stark, D. P., Vidal-García, A., et al. 2017, MNRAS, 472, 2608

    work page 2017

  77. [77]

    & Brinchmann, J

    Shirazi, M. & Brinchmann, J. 2012, MNRAS, 421, 1043

    work page 2012

  78. [78]

    & Wise, J

    Skinner, D. & Wise, J. H. 2020, MNRAS, 492, 4386

    work page 2020

  79. [79]

    2015, ApJ, 808, 139

    Sobral, D., Matthee, J., Darvish, B., et al. 2015, ApJ, 808, 139

    work page 2015

  80. [80]

    & Hernquist, L

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

    work page 2003

Showing first 80 references.