{"id":"954e8012-5499-4d5a-9eb5-75447e3278c8","arxiv_id":"2508.05106","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"The ratio of stalk length to wing span controls whether linden diaspores autorotate, with a threshold near 0.5 doubling flight time by halving descent speed.","lead":"Linden tree seeds tumble and fall faster when the stalk connecting the seed pods to the bract wing is shortened; a stalk at least half as long as the wing span switches the seed into a slow spinning autorotation that doubles flight time. The finding casts the stalk as a functional flight organ, not just a seed holder, and offers a simple control lever for bio-inspired flying devices.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The claimed L/R_S ≈ 0.5 autorotation threshold is set by censored 9 m drops; the short-stalk branch of Fig. 3b may be an experimental-window artifact.","rationale":"The central claim requires that the transition from tumbling to autorotation genuinely diverges as the stalk shortens, with a critical L/R_S of about 0.5. The 9 m tower is the only evidence for the diverging branch; it is a censoring threshold, not a measured onset. This is load-bearing because the hollow markers in Fig. 3b are the points that define the transition: without them, the data only show that longer stalks autorotate sooner, not that there is a sharp threshold. The concern is fixable and does not invalidate the qualitative observation; the paper's other evidence (qualitative stalk-length dependence, PTV vortices, blade-element model) is credible and independently supports the mechanism. The reader's CONDITIONAL verdict remains appropriate; no change to the verdict is needed.","tokens_in":13359,"tokens_out":12927,"duration_ms":149995,"concrete_test":"Repeat the flight experiments for all hollow-marker (no autorotation in 9 m) diaspores from a height of at least 15 m (e.g., a stairwell, crane, or balloon release), recording whether and at what distance autorotation begins; if a substantial fraction of these diaspores transition after 9 m, the L/R_S≈0.5 threshold is an artifact of the experimental window.","verdict_should_be":"UNCHANGED","load_bearing_attack":"In Section III, L_m is defined as the descent distance at which autorotation initiates, but for all cases where no autorotation is seen within the 9 m fall, the authors set L_m = 9 m. These censored, hollow-marker points are exactly the short-stalk (small l) branch that makes L_m/L_N diverge in Fig. 3b, and the divergence is what supports the 'L/R_S larger than about half the span' threshold. The text itself says 'for sufficiently short stalks... we do not observe any autorotation within the experimental window.' Assigning a finite 9 m value to an unobserved event treats a window limit as a physical transition. If a short stalk would autorotate after a longer fall, the threshold would shift or vanish; if it would never autorotate, the L_m values should be infinity, not 9 m. No survival analysis, longer-drop experiment, or model extrapolation is provided to separate these cases. Thus the central quantitative claim is not established at the stated precision.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper investigates the aerodynamic role of the stalk (peduncle) in linden diaspores (Tilia x europaea). Through free-fall experiments on biological and synthetic diaspores with shortened stalks, the authors observe that longer stalks lead to earlier onset of autorotation and a roughly twofold reduction in descent speed. They propose a critical stalk-length-to-wing-span ratio L/R_S ≈ 0.5 for steady autorotation, support this with a data collapse using L_N = U_s^2/g and l/r, and present PTV measurements in water showing leading- and trailing-edge vortices on a curved synthetic wing. The paper also reports that more than 99% of 859 collected diaspores satisfy the proposed threshold, suggesting an evolutionary interpretation.","tokens_in":13661,"tokens_out":6889,"duration_ms":71083,"significance":"The question is significant because it identifies a specific morphological trait—stalk length—as an aerodynamic control element in wind dispersal, with a quantitative design rule that could inform bio-inspired micro-air vehicles. The experimental program is a strength: direct free-fall tests on biological and synthetic diaspores, matched Reynolds and Strouhal numbers, and publicly archived data. The PTV flow measurements provide a plausible physical mechanism (tip-concentrated leading- and trailing-edge vortices). However, the central quantitative threshold is not yet established because it is derived from data censored by the 9 m fall height; the short-stalk branch that produces the divergence in Fig. 3b is precisely the branch where no autorotation was observed within the experimental window. This makes the headline claim conditional on an assumption about unobserved long-fall behavior.","major_comments":[{"comment":"The threshold L/R_S ≈ 0.5 is inferred from the divergence of L_m/L_N for short stalks. In Section III and Fig. 3b, L_m is set to 9 m whenever no autorotation is observed in the experimental window. These are right-censored observations: 'no autorotation in 9 m' is not equivalent to 'no autorotation ever,' and the divergent branch is composed of exactly these hollow markers. Assigning a finite 9 m value to an unobserved event treats the window limit as a physical transition. The Methods also describe experiments at 2 m, 5 m, and 9 m, but the paper does not state which censoring limit applies to each point, even though all hollow markers are assigned 9 m. No survival analysis or longer-fall test is provided. The central quantitative claim therefore rests on an unverified assumption about short-stalk behavior outside the observation window.","section":"Section III, Fig. 3b"},{"comment":"The collapse in Fig. 3b uses the normalization L_N = U_s^2/g, where U_s is the steady descent velocity of the same diaspore with its longest stalk. Because U_s is itself an outcome measured from the same flight experiments, the normalization is not independent of the data being collapsed. The authors should demonstrate that the collapse is robust when U_s is estimated from an independent relation such as Eq. (A1) (weight/area scaling) or from a fixed reference stalk length; otherwise part of the apparent universality of the curve may be introduced by the normalization. This is secondary to the censoring issue but affects the quantitative form of the proposed design rule.","section":"Section III, definition of L_N"},{"comment":"The stated result is a threshold in L/R_S, but the data collapse and divergence are plotted against l/r, where l is the distance between the wing and the center of mass and r is the horizontal distance defined in Fig. 2b. These ratios are not identical, and the text does not explicitly derive the mapping from l/r > 1 to L/R_S ≈ 0.5. Please state the relationship used (e.g., r ≈ R_S/2 for the tested geometries) and show that the threshold is insensitive to this mapping; as written, the reader cannot reproduce the conversion from the plotted variable to the headline claim.","section":"Section III, Figs. 2b and 3b"}],"minor_comments":[{"comment":"The last sentence says 'For longer stalks (c) and (d)' but the context requires 'shorter stalks'; the wording is inconsistent with the preceding sentence.","section":"Fig. 5 caption"},{"comment":"The drag force expression repeats C_L(alpha); it should be C_D(alpha).","section":"Eq. (A4)"},{"comment":"'decent height' should be 'descent height' in the two places where the 9 m experimental window is described.","section":"Section III and Fig. 3b caption"},{"comment":"The log-normal fit parameters are given, but no goodness-of-fit measure or uncertainty is reported; please add a quantitative assessment of the fit.","section":"Fig. 1d"},{"comment":"The Methods state that experiments were performed at 2 m, 5 m, and 9 m fall heights, but it is not specified which data points in Figs. 3a/b come from which height. This is important for interpreting the censoring limit and should be clarified.","section":"Appendix A.3"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is within scope for physics.bio-ph and the combination of direct flight experiments, synthetic-diaspore controls, and matched-Reynolds flow measurements is commendable. The main obstacle is the censored-data inference of the L/R_S ≈ 0.5 threshold; if the authors can provide a survival analysis, longer-fall data, or a model that predicts the onset distance, the paper would be substantially stronger."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Two things to know. First, the paper's main qualitative result is real: for linden diaspores, a longer stalk makes the fruit start autorotating sooner, and autorotation roughly halves descent speed. Second, the advertised quantitative threshold—stalk length about half the wing span—is not supported by the data as presented, because the short-stalk branch of the collapse is censored at the 9 m drop height.\n\nWhat is new and good. Previous work on autorotating seeds concentrated on samaras and on mass distribution or wing curvature. Nobody had identified the stalk-to-span ratio as the controlling parameter for Tilia diaspores. The experimental setup is thoughtful: biological diaspores with cut stalks, synthetic replicas where shape and weight are controlled, and PTV flow measurements in water at matched Reynolds and Strouhal numbers. The flow visualization showing a leading-edge vortex and a smaller trailing-edge vortex near the wing tip is a solid piece of evidence that the curved wing generates lift the way the authors think. The collection of 859 field diaspores and the log-normal distribution of L/R_S is careful descriptive work.\n\nWhere it gets soft. The threshold L/R_S ≈ 0.5 is read from a data collapse, Fig. 3b, where L_m is the distance to autorotation onset. But for every diaspore that never autorotated within the 9 m tower, the authors set L_m = 9 m. Those censored points are exactly the short-stalk branch that makes L_m/L_N diverge. That divergence is an artifact of the experimental window unless you have independent evidence that these diaspores would never autorotate no matter how far they fell. The paper offers no such evidence—no longer drops, no survival analysis, no theoretical argument for a hard cutoff. So the 'critical stalk length' is not established at the stated precision. The qualitative statement that shorter stalks take longer or never autorotate within the window is fine, but the specific 0.5 ratio should be treated as provisional. There are also minor internal text slips: the caption of Fig. 2 says 'longer stalks' where it should say 'shorter stalks' for (c) and (d), and the divergence direction in the text is sometimes described loosely. The figure has no error bars on the collapsed data.\n\nBottom line. This paper deserves a serious referee. The core observation is likely right and the experimental methods are well above average. The censoring issue is fixable—either with longer-drop experiments, an explicit censored-data model, or a more cautious claim—and the text needs a round of corrections. I'd send it out, but the referee should not accept the 0.5 threshold as established.\n\nFor my own work: I'd cite the qualitative finding if I were in seed dispersal or bio-inspired flight, but I wouldn't rely on the threshold. Reading group: maybe worth a discussion on censoring in experimental science.","headline":"The qualitative finding—longer stalks speed up autorotation—is solid, but the L/R_S≈0.5 threshold is not established because the short-stalk branch is censored at the 9 m drop height.","tokens_in":14088,"tokens_out":2981,"would_cite":true,"duration_ms":32921,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Linden diaspores use their stalk as an aerodynamic lever: steady autorotation requires the stalk to exceed about half the wing span, roughly doubling flight time compared with tumbling.","keywords":["linden diaspore","Tilia x europaea","autorotation","wind dispersal","stalk length","wing span","leading-edge vortex","blade element theory"],"falsifier":"Drop synthetic and natural linden diaspores with stalk-to-span ratios below 0.5 from a height of 30 m or more in still air, for example from a crane or tall building, and record the full trajectory. If any of them reach steady autorotation after falling farther than 9 m, the threshold is an artifact of the drop window; if none do even after 30 m, the threshold is a physical transition.","tokens_in":13260,"feed_emoji":"🍃","tokens_out":7524,"duration_ms":81298,"temperature":0.7,"pith_summary":"The paper asks why linden diaspores carry their heavy seed pods at the end of a long stalk, and answers that the stalk is part of the flight machinery. By dropping real and synthetic diaspores with shortened stalks, it shows that the ratio of stalk length to wing span controls whether the seed settles into steady autorotation or keeps tumbling unsteadily. Steady autorotation halves descent speed and therefore roughly doubles flight time, so stalk length should matter for wind dispersal. The paper also images the flow around a model wing and finds a leading-edge vortex and a trailing-edge vortex near the wing tip that generate extra lift. It concludes that stalk length is likely a selected trait, since more than 99% of collected natural diaspores exceed the critical stalk-to-span ratio.","feed_headline":"Stalk length decides whether linden seeds spin or tumble","feed_subtitle":"Seeds with stalks shorter than half the wing span tumble; longer stalks trigger autorotation and halve the descent speed.","key_machinery":"The stalk acts as a lever arm that controls the torque balance between the aerodynamic forces on the curved bract and the weight of the pods. For the same forces and weight, a shorter stalk produces a larger torque about the tumbling axis, so the tumbling motion persists to larger angles and longer fall distances before the spin can take over. The quantitative argument is blade-element theory: the bract is divided into thin slices, each with its own local lift and drag, and the equilibrium descent speed $U$, rotation speed $\\Omega$, and stalk tilt are found where the vertical force, spin torque, and tilt torque all vanish. The flow visualization supplies the lift source, a leading-edge vorte","core_discovery":"The central claim is that the peduncle of a linden diaspore ($Tilia\\ x\\ europaea$) is an aerodynamic control element, not just a mechanical link, and its length sets the transition between unsteady tumbling and steady autorotation. In still-air drops of biological and synthetic diaspores, the fall distance $L_m$ needed before the spin sets in shrinks as the stalk lengthens and diverges for short stalks; the data collapse when $L_m$ is normalized by the length scale $L_N=U_s^2/g$ and plotted against $l/r$, the ratio of center-of-mass offset to the wing lever arm. The criterion is that steady autorotation requires the stalk to be longer than about half the wing span, $L/R_S\\gtrsim 0.5$, and mo","pith_inferences":["If the 9 m drop height is the censoring limit, the true threshold could be lower than $L/R_S\\approx 0.5$; a longer-drop test would tell whether the threshold is physical or an artifact of the observation window.","The same lever-arm idea suggests a passive control strategy for micro-air vehicles: moving the payload along a stalk switches the flight mode between fast tumbling and slow autorotation without altering the wing.","The log-normal distribution of $L/R_S$ across 859 sampled diaspores hints at selection, but a direct test would compare dispersal distances of short- versus long-stalked diaspores or compare related Tilia species with different stalk geometries."],"forward_implications":["A diaspore with a natural long stalk starts autorotating almost immediately after release, so it spends nearly its whole fall in the slow spinning state rather than in fast tumbling.","Shortening the stalk moves the onset of autorotation past the height of a typical tree, so a slightly shorter stalk can turn a dispersing seed into a rapidly falling one.","Since autorotation cuts descent speed by about a factor of two, stalk length translates directly into flight time and therefore into potential wind-transport distance.","The effect is reproducible with synthetic diaspores of different wing geometries and weights, so the criterion is mechanical and geometric, not a peculiarity of one tree.","The blade-element model reproduces the plateau and divergence of the onset distance, giving a predictive tool for when a stalked-wing seed will autorotate."],"supporting_citations":[{"why":"Supplies the reference flight data for Tilia miqueliana (descent and rotation speeds) and the classical framework for autorotating seed flight.","marker":"[1]"},{"why":"Provides the kinematics and torque picture of falling maple seeds' transition from tumbling to helical autorotation that the stalk-lever argument extends.","marker":"[6]"},{"why":"Identifies the leading-edge vortex as the source of elevated lift on rotating wings, the mechanism the flow measurements confirm for linden.","marker":"[15]"},{"why":"Provides the synthetic whirling-fruit approach and the result that curvature optimizes lift, used to design the synthetic diaspores.","marker":"[17]"},{"why":"Gives the classical scaling descent speed $U\\propto\\sqrt{mg/\\rho A}$ for autorotating seeds used for the steady-state balance.","marker":"[20]"},{"why":"Defines the autorotation boundary concept that the stalk-length criterion is a specific instance of.","marker":"[22]"},{"why":"Supports the premise that changing mass distribution can suppress autorotation, motivating the stalk-length experiments.","marker":"[23]"},{"why":"Supplies the blade-element lift and drag coefficients and camber correction used for the equilibrium calculations.","marker":"[32]"}],"fun_headline_variants":["Linden seed stalk length flips flight from tumble to spin","Stalk-to-wing ratio decides linden seed autorotation","Longer linden stalks halve seed descent via autorotation","Seed stalk sets spin: linden diaspore flight secret","Half-wing stalk triggers linden seed autorotation"],"cache_read_input_tokens":2816,"weakest_assumption_plain":"The inference of a critical stalk length rests on treating the 9 m drop tower as an infinite falling distance: seeds that have not started autorotating after 9 m are classed as non-rotators, so the claimed threshold around $L/R_S=0.5$ could be set by the experimental window rather than by physics.","fun_headline_variants_meta":{"raw":{"variants":["Linden seed stalk length flips flight from tumble to spin","Stalk-to-wing ratio decides linden seed autorotation","Longer linden stalks halve seed descent via autorotation","Seed stalk sets spin: linden diaspore flight secret","Half-wing stalk triggers linden seed autorotation"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000164,"raw_usage":{"total_tokens":1079,"prompt_tokens":734,"completion_tokens":345,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":478,"completion_tokens_details":{"reasoning_tokens":262}},"tokens_in":478,"tokens_out":345,"duration_ms":3673,"temperature":1.0,"reasoning_tokens":262,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-05T23:32:33.896344+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Drop synthetic and natural linden diaspores with stalk-to-span ratios below 0.5 from a height of 30 m or more in still air, for example from a crane or tall building, and record the full trajectory. If any of them reach steady autorotation after falling farther than 9 m, the threshold is an artifact of the drop window; if none do even after 30 m, the threshold is a physical transition.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the reference flight data for Tilia miqueliana (descent and rotation speeds) and the classical framework for autorotating seed flight."},{"cited_title":"From𝑈andΩ, we get𝑅𝑒and𝑆𝑡for diaspores of different weight and different geometry","cited_arxiv_id":null,"evidence_quote":"Provides the kinematics and torque picture of falling maple seeds' transition from tumbling to helical autorotation that the stalk-lever argument extends."},{"cited_title":"De Langre, Effects of wind on plants, Annu","cited_arxiv_id":null,"evidence_quote":"Identifies the leading-edge vortex as the source of elevated lift on rotating wings, the mechanism the flow measurements confirm for linden."},{"cited_title":"Vogel, Drag and reconfiguration of broad leaves in high winds, Journal of Experimental Botany40, 941 (1989)","cited_arxiv_id":null,"evidence_quote":"Provides the synthetic whirling-fruit approach and the result that curvature optimizes lift, used to design the synthetic diaspores."},{"cited_title":"Lentink and M","cited_arxiv_id":null,"evidence_quote":"Defines the autorotation boundary concept that the stalk-length criterion is a specific instance of."},{"cited_title":"Azuma and Y","cited_arxiv_id":null,"evidence_quote":"Supports the premise that changing mass distribution can suppress autorotation, motivating the stalk-length experiments."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the blade-element lift and drag coefficients and camber correction used for the equilibrium calculations."}],"review_version":1}