{"id":"37776e47-bcba-4612-92cd-9dc82f4fee7d","arxiv_id":"1909.01011","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"A 350 micron census of 575 massive clumps finds that the mass of the most massive embedded core is about 30% of the clump mass, and identifies 27 massive quiescent core candidates.","lead":"An APEX/SABOCA 350 micron survey of 575 massive clumps resolves 1120 compact cores and maps how fragmentation depends on clump properties. The paper reports that the most massive core in a clump scales nearly linearly with total clump mass, and it isolates 27 massive cores with no infrared signposts of star formation.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The headline M_mmf–M_clump correlation is not separated from the common d^2 scaling of both masses (Eq. 2) or from the built-in 'largest sub-region of parent' effect; the self-gravity interpretation is therefore not yet supported.","rationale":"The most load-bearing concern is the abstract's first headline result: the strong M_mmf-M_clump correlation and the inference that self-gravity plays an important role. The quiescent-core sample is also a headline, but the authors explicitly frame those 27 objects as 'promising candidates ... or deeply embedded high-mass protostars' and report that ALMA follow-up of two candidates reveals protostars; that claim is already hedged and the paper calls for high-resolution confirmation. The correlation, by contrast, is asserted as a strong, unhedged statistical result. The weakness is not that the correlation is absent; it is that the presented statistics cannot distinguish a physical mass-mass relation from the common d^2 scaling of both masses via Eq. (2) within the 2-4 kpc sample, nor from the fact that the most massive fragment is necessarily a sub-region of the clump, so size-of-sample and aperture effects will produce a high correlation for almost any fragmentation process. This is a correctness risk in the central physical inference, not a disagreement with consensus. The paper deserves credit for a careful census, the stated factor 2-3 mass uncertainties, and an independent dendrogram cross-check; however, the dendrogram check uses the same maps and the same d^2 mass formula, so it does not remove the confound. The reader identified the quiescence classification as the weakest assumption; that concern is real but largely acknowledged by the authors. Because the reader's rationale already mentions the partly built-in nature of the correlation, my agreement is partial. The appropriate outcome remains conditional acceptance subject to the distance-control test: if the test shows the correlation collapses, the self-gravity interpretation must be removed or heavily qualified, while the catalog and core sample would remain valuable.","tokens_in":45457,"tokens_out":7830,"duration_ms":81779,"concrete_test":"Re-fit the M_mmf-M_clump relation within narrow distance bins (e.g., 2-2.5, 2.5-3, 3-3.5, 3.5-4 kpc) and compute the partial Spearman correlation between log M_mmf and log M_clump controlling for log d; also regress log(M_mmf/M_clump) against log d. If within-bin Spearman drops well below 0.92, or if the residual mass ratio scales as roughly d^2, the headline correlation is dominated by the common distance factor rather than by clump self-gravity.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Sec. 5.3.1 and Fig. 15 present the strongest physical claim: log M_mmf = 0.96 log M_clump - 1.18 with Spearman rho = 0.92, interpreted as evidence that self-gravity sets the clump-to-core mass relation. Two features of the measurement make this interpretation insecure. First, Eq. (2) gives both M_clump and M_mmf as M = mu m_H N(H2) A d^2. The 2-4 kpc subsample still spans a factor ~2 in distance, so both masses share a strong common d^2 scaling; two quantities that both scale as d^2 will produce a tight log-log relation with slope near 1 even if no physical mass-mass relation exists. The Spearman coefficient is rank-based and does not remove this. Second, M_mmf is by construction the largest sub-structure inside M_clump. Size-of-sample and aperture effects (more massive clumps have more resolved sub-structures, and fragment apertures scale with distance and resolution) generically produce M_mmf proportional to M_clump. The dendrogram cross-check in App. C reproduces the correlation, but it inherits the same distance scaling and subset relation, so it does not constitute a null test. The paper never reports a partial correlation controlling for distance, a narrow-distance-bin fit, or a null model in which clump mass is randomly partitioned into resolution elements. Without such a test, the observed slope ~0.96 and rho ~0.92 cannot uniquely support the self-gravity interpretation.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper reports SABOCA 350 μm observations of 204 ATLASGAL massive clumps, extracts 1120 compact sources with Gaussclumps, and derives clump and core physical properties from pixel-by-pixel SED fitting using Herschel, LABOCA, and SABOCA data. The authors classify cores as star-forming or quiescent using 22/24 μm and 70 μm point-source associations, analyze fragmentation levels as a function of clump properties, and identify 27 massive (>100 M_sun) quiescent cores within 5 kpc as candidate massive pre-stellar cores or deeply embedded high-mass protostars. The main quantitative claims are a moderate correlation of fragment number with clump density and Jeans number, and a strong correlation between the most massive fragment mass and clump mass, log M_mmf = 0.96 log M_clump - 1.18 (Spearman rho = 0.92), which the authors interpret as evidence that self-gravity plays an important role in clump-scale structure formation.","tokens_in":45823,"tokens_out":6493,"duration_ms":67321,"significance":"The paper delivers a valuable observational dataset: a homogeneous 350 μm survey of 204 clumps, SED-based masses and luminosities for 971 cores, and a public machine-readable catalog. The empirical correlations are benchmarked against an independent dendrogram extraction in Appendix C, which strengthens the credibility of the source-extraction-dependent results. The 27 massive quiescent cores constitute a useful target list for high-resolution follow-up with ALMA. However, the physical interpretation of the headline M_mmf–M_clump correlation as evidence for self-gravity is not uniquely supported by the present analysis, because both masses share a common d^2 scaling and the most massive fragment is by construction a sub-structure of the clump. The quiescent classification of the massive cores also carries a contamination risk that the paper itself partly documents. These issues are addressable with additional null tests and more cautious wording, so the paper is suitable for major revision rather than rejection.","major_comments":[{"comment":"The headline correlation log M_mmf = 0.96 log M_clump - 1.18 with Spearman rho = 0.92 is interpreted as evidence that self-gravity sets the clump-to-core mass relation, but two built-in effects are not separated. First, Eq. (2) gives both M_clump and M_mmf as M = mu m_H N(H2) A d^2, so even within the 2-4 kpc distance bin both quantities share a common d^2 scaling; a factor of 2 in distance produces a spread of ~0.6 dex in log d^2 and will push a log-log correlation toward a slope near unity even if no intrinsic mass-mass relation exists. The Spearman coefficient is rank-based and does not remove this common scaling. Second, M_mmf is by construction the largest sub-structure inside M_clump; Appendix C itself notes that the relative size difference between parent and child structures 'together lead again to the tight correlation between the mass of the most massive fragment and the parental clump mass,' illustrating the definitional component. The dendrogram cross-check in App. C inherits both the distance scaling and the subset relation, so it does not serve as a null test. To support the self-gravity interpretation, please report a partial Spearman correlation controlling for distance, a narrow-distance-bin fit (e.g., 2.5-3.5 kpc), or a null model in which fragment masses are randomly assigned to clumps of the observed mass function. Without such a test, the claim should be stated as a correlation whose physical origin requires further investigation, not as direct evidence for self-gravity.","section":"Sec. 5.3.1, Fig. 15, Eq. (2), App. C"},{"comment":"The identification of 27 massive quiescent cores as candidates for massive pre-stellar cores rests on the absence of compact 24/22 μm and 70 μm counterparts. This classification criterion is not sufficient to exclude deeply embedded high-mass protostars, which can be invisible at these wavelengths; the paper itself notes in Sec. 5.4 that deeply embedded high-mass protostars have been detected toward mid-infrared quiet massive cores. The ALMA follow-up of two candidates, G333.1298-0.5602 and G333.4659-0.1641, shows that the brightest sub-arcsecond fragments are high-mass protostellar objects rather than pre-stellar cores. If a comparable fraction of the remaining candidates are protostellar, the claim of a population of massive pre-stellar cores would be substantially weakened. Please quantify the possible contamination using the derived 70 μm opacities and the sensitivity limits of the mid-infrared catalogs, clearly mark the ALMA-confirmed sources in Table 5, and adjust the strength of the conclusion accordingly.","section":"Sec. 4.4 and Sec. 5.4, Table 5"}],"minor_comments":[{"comment":"The fitted relation is written as 'M_fragments = 0.96 log M_clumps - 1.18'; since the horizontal axis is logarithmic, the left-hand side should be log M_mmf (and similarly in Fig. C.3).","section":"Fig. 15 and Fig. C.3 captions"},{"comment":"The quantity Nmm is used repeatedly but never defined; please define the acronym or replace it with a descriptive symbol such as N_frag.","section":"Sec. 5.3.1 and Fig. 14"},{"comment":"The symbols in the SiO and HCO+/H13CO+ columns (a, b, c, d, e, y, '-') are only partially explained in the table notes; please give a complete legend so the table can stand alone.","section":"Table 5"},{"comment":"The statement that 67% of fields exhibit filamentary(-like) morphology is based on visual inspection; please state this explicitly and, if possible, provide a simple quantitative criterion or acknowledge that the fraction is a visual estimate.","section":"Sec. 3.1"}],"recommendation":"major_revision","confidential_remarks":"The manuscript fits the scope of Astronomy & Astrophysics and the dataset is genuinely useful. The referee report focuses on two improvements that I consider necessary before publication: a null test for the common distance/subset origin of the M_mmf–M_clump correlation, and a more quantitative treatment of contamination in the quiescent-core sample. The authors' own Appendix C and Sec. 5.4 caveats show that these concerns are not manufactured; they need to be addressed rather than only cited."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Let me give you the short version: this is a careful, useful census with a genuinely new target list, but the headline interpretation of the M_mmf–M_clump correlation is shakier than the abstract lets on.\n\nWhat is actually new: the sample is roughly twice the size of the earlier SHARC-II work (Merello et al. 2015), they extract 1120 compact sources, and they provide a clean list of 27 massive quiescent-core candidates for ALMA follow-up. The M_mmf–M_clump relation, M_mmf = 0.31 M_clump^0.96, is new, and the dendrogram comparison in App. C is a legitimate robustness check. The paper is also transparent about its main systematics: fixed dust opacity and beta give factor 2–3 mass uncertainties, distance-dependent resolution is discussed, and the KS tests on classification biases are honest. The citation pattern is fine; the relevant earlier 350 um study and the hierarchical-collapse literature are engaged rather than ignored.\n\nThe soft spots are moderate, not fatal. First, the strong M_mmf–M_clump correlation has a built-in component: the most massive fragment is a subregion of the clump, and both masses carry a common d^2 distance term. Restricting to 2–4 kpc helps but does not remove it; they never run a partial correlation on distance or a null model with randomly partitioned clump mass. The dendrogram cross-check inherits the same distance scaling and parent–child relation, so it is not a null test. The empirical relation is probably real, but the self-gravity interpretation needs support from such a control. Second, the quiescence classification rests on the absence of compact 24/70 um emission. The paper itself notes that ALMA follow-up of two candidates found high-mass protostellar fragments, which means some of the 27 are protostellar. They do call them candidates and hedge in the abstract, but 'robust' in the conclusion overstates it. Third, all mass values carry the acknowledged factor 2–3 opacity uncertainty.\n\nWho it is for: people working on massive star formation fragmentation, and especially anyone building ALMA follow-up samples. It deserves a serious referee. My recommendation: send it to review, and ask the authors for a distance-controlled version of the M_mmf–M_clump analysis and softer wording on the pre-stellar candidates.","headline":"A careful, useful census with a genuinely new target list, but the headline self-gravity interpretation of the M_mmf–M_clump correlation is shakier than the abstract lets on.","tokens_in":46353,"tokens_out":4934,"would_cite":true,"duration_ms":54780,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"The most massive fragment inside a clump scales almost one-to-one with clump mass, a sign that self-gravity shapes small-scale structure.","keywords":["massive star formation","clump fragmentation","pre-stellar cores","submillimeter observations","SABOCA 350 micron","ATLASGAL clumps","Jeans fragmentation","filamentary structure"],"falsifier":"Image the 27 massive quiescent cores with sub-arcsecond interferometry and search for outflow wings, hot-core tracers, or compact continuum sources; widespread detections would falsify the pre-stellar interpretation. A complementary test would re-derive the $M_{\\rm mmf}$-$M_{\\rm clump}$ correlation after convolving every map to a common physical resolution, since a slope well below 0.96 would weaken the self-gravity claim.","tokens_in":45268,"feed_emoji":"🌟","tokens_out":11384,"duration_ms":107836,"temperature":0.7,"pith_summary":"This paper asks what sets the sub-parsec structure of massive star-forming clumps: why some clumps break into a few heavy cores while others fragment into many light ones. Using APEX/SABOCA 350 μm maps at 8.5 arcsecond resolution toward more than 200 ATLASGAL clumps, the authors resolve 1120 compact sources and derive dust temperatures, masses, and luminosities for 971 of them through pixel-by-pixel SED fitting with Herschel and ATLASGAL data. They find that the mass of the most massive fragment tracks the parent clump mass almost one-to-one, with $\\log M_{\\rm mmf} = 0.96 \\log M_{\\rm clump} - 1.18$ (about 30 percent of the clump mass), a strong correlation they read as evidence that clump self-gravity governs core-scale structure. They also report that fragment number correlates only moderately with gas density and with the Jeans-fragmentation prediction, and that fragments are 10-100 times the thermal Jeans mass. Finally, they identify 27 massive quiescent cores above 100 solar masses within 5 kpc as the best current candidates for massive pre-stellar cores, while cautioning that deeply embedded protostars could masquerade as quiescent.","feed_headline":"Most massive fragment tracks clump mass in 350-micron survey","feed_subtitle":"Heaviest fragments hold about 30 percent of clump mass, and 27 quiet cores are pre-stellar candidates.","key_machinery":"The load-bearing object is the empirical relation between the most massive fragment mass and the clump mass, $M_{\\rm mmf} = 0.31\\,M_{\\rm clump}^{0.96}$, together with the data chain that produces it. That chain consists of SABOCA 350 μm maps at 8.5 arcsecond resolution, Fourier-combined with Herschel SPIRE 350 μm data to recover extended emission; pixel-by-pixel modified-blackbody SED fits at 10 arcsecond resolution that convert 70-870 μm fluxes into dust temperature, column density, and mass; Gaussclumps, a Gaussian-decomposition source extractor, to define the compact cores; and positional matching to 22-24 μm and 70 μm point-source catalogs to separate star-forming from quiescent cores. The relation carries the self-gravity argument because it ties the parent reservoir to its most massive product across two orders of magnitude in clump mass.","core_discovery":"On the paper's own terms, the central discovery is a scaling law connecting the two ends of the fragmentation hierarchy: inside massive clumps, the most massive fragment obeys $M_{\\rm mmf} = 0.31\\,M_{\\rm clump}^{0.96}$ in the 2-4 kpc distance-limited sample (Spearman coefficient 0.92, p < 0.001), meaning roughly 30 percent of the clump mass is already assembled in its single heaviest core. The same analysis finds that the number of fragments correlates only moderately with clump density (Spearman ~0.41) and with the number of thermal Jeans masses (0.31-0.40), and that typical fragment masses exceed the thermal Jeans mass by 10-100 times. The companion result is a census of 27 quiescent cores with $M > 100\\,M_\\odot$ at distances below 5 kpc that show no compact 22-24 μm or 70 μm counterparts; the paper presents them as promising candidates for massive pre-stellar cores, while explicitly leaving open the alternative that they are deeply embedded high-mass protostars.","pith_inferences":["If the $M_{\\rm mmf}$-$M_{\\rm clump}$ relation holds at fixed physical resolution, it would give observers a predictor: the most massive star in a young cluster could be anticipated from the clump mass, much as the $m_{\\rm max}$-$M_{\\rm ecl}$ relation anticipates the most massive star from cluster mass; the paper compares the two but does not claim equality.","Because the sample is selected to be weak or dark at 24 μm, the correlations are measured on an early-stage population; extending the same analysis to more evolved, mid-infrared-bright clumps would show whether the 30 percent fraction changes with evolution.","Smoothing all maps to a common physical resolution before measuring $M_{\\rm mmf}$ would test whether part of the correlation is simply a distance and resolution artifact, since more distant clumps are seen with worse linear resolution."],"forward_implications":["Because the most massive fragment holds roughly 30 percent of the clump mass, the clump mass sets the upper end of the core mass function and, plausibly, the mass of the cluster that forms.","The scarcity of massive quiescent cores above 100 solar masses (27 in the sample) implies that such dense structures either collapse quickly or quickly become visible in the mid-infrared.","Fragment spacing is broadly consistent with thermal Jeans lengths, but individual fragments are 10 to 100 times the thermal Jeans mass, so turbulent or self-gravitating support must set the fragment masses.","Quiescent and star-forming cores have similar masses but different surface densities and temperatures, supporting a picture where the quiescent cores are the same objects at an earlier stage along one evolutionary path.","The steepening of the mass-radius relation toward smaller scales (slope near 3 rather than 2) points to self-gravity becoming dominant from clump to core scales."],"supporting_citations":[{"why":"Provides the ATLASGAL 870 μm survey from which the parent clump sample is selected.","marker":"Schuller et al. 2009"},{"why":"Supplies the ATLASGAL compact-source catalog, clump sizes, and the target list for SABOCA follow-up.","marker":"Csengeri et al. 2014"},{"why":"Provides kinematic distances and physical parameters for the clumps that define the distance-limited samples.","marker":"Urquhart et al. 2018"},{"why":"Supplies the Hi-GAL PACS/SPIRE maps and the 70 μm point-source catalog used for SED fitting and activity classification.","marker":"Molinari et al. 2016"},{"why":"Provides the MIPSGAL 24 μm point-source catalog used to flag star-forming cores.","marker":"Gutermuth & Heyer 2015"},{"why":"Is the earlier 350 μm fragmentation study that this survey doubles in sample size and extends in parameter space.","marker":"Merello et al. 2015"},{"why":"Establishes the quiescent versus star-forming core classification and the use of SiO line-wings to probe embedded activity.","marker":"Motte et al. 2007"},{"why":"Reports ALMA follow-up showing two of the quiescent candidates are actually high-mass protostars, framing the interpretation of the candidate sample.","marker":"Csengeri et al. 2017b"}],"fun_headline_variants":["Most massive fragment scales with clump mass in 350-micron survey","Fragments trace clump mass: heaviest core holds ~30%","27 quiet cores might be massive pre-stellar seeds","Massive clump census flags pre-stellar core candidates","Fragmentation stats: heaviest core scales with clump mass"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The 27 pre-stellar candidates are defined by the absence of a compact 24 or 70 micron infrared source, yet a deeply embedded massive protostar can be hidden at exactly those wavelengths, so a core without such a counterpart is not guaranteed to be starless.","fun_headline_variants_meta":{"raw":{"variants":["Most massive fragment scales with clump mass in 350-micron survey","Fragments trace clump mass: heaviest core holds ~30%","27 quiet cores might be massive pre-stellar seeds","Massive clump census flags pre-stellar core candidates","Fragmentation stats: heaviest core scales with clump mass"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000894,"raw_usage":{"total_tokens":3935,"prompt_tokens":1108,"completion_tokens":2827,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":724,"completion_tokens_details":{"reasoning_tokens":2737}},"tokens_in":724,"tokens_out":2827,"duration_ms":20726,"temperature":1.0,"reasoning_tokens":2737,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T05:29:22.547491+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Image the 27 massive quiescent cores with sub-arcsecond interferometry and search for outflow wings, hot-core tracers, or compact continuum sources; widespread detections would falsify the pre-stellar interpretation. A complementary test would re-derive the $M_{\\rm mmf}$-$M_{\\rm clump}$ correlation after convolving every map to a common physical resolution, since a slope well below 0.96 would weaken the self-gravity claim.","supporting_citations":[],"review_version":1}