henhaobio
Sign in

Biology LettersPaleontologyPeer-reviewed journal article

Predator-prey scaling laws support a suspension-feeding lifestyle in Cambrian luolishaniid lobopodians

Jared RichardsiD, Javier Ortega‐HernándeziD

PDFDOIPreprint
In-depth full-text digest available · Jump to digest ↓

Abstract

The early Paleozoic saw a dramatic diversification of shelly epibenthic metazoans adapted to suspension and filter-feeding, but the extent to which these radiations affected the evolution of non-biomineralized suspension-feeding taxa is uncertain because these organisms are not typically well represented in the fossil record. Luolishaniids are a highly derived and disparate clade of (typically) armoured lobopodians widely interpreted as having a suspension-feeding ecology based on the presence of five or six anterior pairs of setulose appendages. Luolishaniids are globally widespread and represent the only Cambrian non-biomineralized free living epibenthic bilaterians regarded as having a suspension feeding mode of life, but their proposed ecology only relies on a qualitative interpretation of their functional morphology. Here we test the hypothesis that the setulose appendages of luolishaniids were adapted for a suspension feeding function. Quantitative morphological comparisons reveal a positive and statistically significant relationship between body length and the mesh spacing of the setulose anterior limbs of luolishaniids. This pattern correlates with suspension-feeding scaling in extant aquatic organisms, and suggests that luolishaniids primarily fed on mesoplanktonic seston. We provide quantitative evidence for suspension-feeding in luolishaniids, which represents the first statistically supported example of modern-like predator-prey scaling patterns observed in Cambrian soft-bodied metazoans.

In-depth summary

Claude Opus 5.5 · xhigh effort · 8 Nov 2026 · from the full text of version 4 (4 Mar 2026) · CC BY 4.0

Cambrian luolishaniid lobopodian appendage mesh spacing scales sublinearly with body length, and their predator-prey size ratios match modern suspension feeders rather than raptorial chaetognaths.

Soft-bodied suspension feeders are rarely fossilized, and the suspension-feeding interpretation of luolishaniid lobopodians rested only on qualitative reading of their setulose anterior limbs, leaving raptorial or sediment-sieving alternatives untested. This work tests it via mesh-spacing allometry and predator-prey size scaling against modern invertebrates, giving the first quantitative support for modern-like suspension-feeder scaling in Cambrian soft-bodied animals.

Key results

  1. SuggestedSetal mesh spacing on luolishaniid anterior appendages rises sublinearly with genus body length on the log scale, a pattern consistent with allometry in extant suspension-feeding aquatic invertebrates.

    Mixed model with genus random effect: F(1, 4.82) = 8.52, p = 3.45×10^-2, marginal R^2 = 0.35, conditional R^2 = 0.65, slope 0.36; OLS on genus means: F(1,5) = 7.72, p = 3.91×10^-2, R^2 = 0.61, slope 0.37, intercept 1.72; Harvey-Collier 0.36 (df 4, p = 0.74) supports log-scale linearity. · Luolishaniid Allometry and Prey Size Estimate; Data Analysis

  2. ShownWhen converted to ESD (diameter of a sphere of equal volume), modeled luolishaniid predator and prey sizes fall on one strong log-log scaling line with modern suspension-feeding invertebrates. Chaetognaths were left out of the fit.

    OLS on log scale: F(1,30) = 198.1, p = 9.40×10^-15, R^2 = 0.87; Log10(Ŷ) = 0.90 × Log10(X) - 1.13; Harvey-Collier 0.91 (df 29, p = 0.37), so the linearity assumption held. · Luolishaniid Macroecology and Modern Suspension…; Data Analysis

  3. SuggestedPredator-to-prey ESD ratios set luolishaniids apart from raptorial chaetognaths and group them with filter-feeding clades. The authors use this to reject a raptorial feeding mode.

    Chaetognaths average about five times their prey's size; copepods and rotifers about 20 times; luolishaniids, trochozoans and cladocerans roughly 25–50. Luolishaniid appendages would retain seston about 30–35 times smaller than body size in ESD terms, similar to larval mollusks and cladocerans. · Luolishaniid Macroecology and Modern Suspension…; Luolishaniid Macroecological Scaling and…; Tables

  4. SuggestedPrey sizes estimated from mesh spacing make most luolishaniids mesoplankton feeders. Only the smallest-bodied taxa (Luolishania and, per the Discussion, Ovatiovermis) could reach larger microplankton, and zooplankton larvae are proposed as the main prey.

    Mesh 108.4 µm in Luolishania (14 mm body) and 161.1 µm in Ovatiovermis (18 mm); ca. 200 to 450 µm from Collinsovermis (35 mm) to Acinocricus (96 mm). Minimum prey ~230 to 460 µm for most taxa vs ~150 µm for Luolishania. Cambrian phytoplankton is typically below 100 μm. · Luolishaniid Allometry and Prey Size Estimate; Luolishaniid Macroecological Scaling and…; Tables

  5. SuggestedLuolishaniid prey-size estimates overlap those of suspension-feeding radiodonts, even though radiodont bodies are far larger. This suggests mesoplankton and large microplankton were common food for several Cambrian to Early Ordovician suspension-feeding clades.

    Overlap with Aegirocassis benmoulai (ca. 200 cm trunk), Pahvantia hastata (ca. 24 cm estimated) and Tamisiocaris borealis (ca. 34 cm estimated), while luolishaniids are roughly 10 to 100 mm long. The overlap is shown graphically, with no formal statistical test. · Luolishaniid Macroecological Scaling and…; INTRODUCTION; Figure legends

Methods, data and reproducibility

Setal mesh spacing on anterior limbs, measured from images of all described luolishaniids, was regressed on log body length (mixed model with genus random effect; OLS on means). Trunks modeled as cylinders were converted to equivalent spherical distance (ESD) and regressed against minimum estimated prey ESD, together with literature data on modern suspension feeders.

  • High-resolution published images plus maximum trunk length and width from original descriptions for all previously described luolishaniid taxa · luolishaniid lobopodians (extinct, Cambrian)
  • Unnamed Emu Bay Shale luolishaniid; body length extrapolated from the fitted power law and excluded from the main analyses · luolishaniid lobopodian (extinct, Cambrian) · n = single incomplete specimen
  • Literature-compiled predator and prey ESD values for modern suspension-feeding marine invertebrates (e.g., copepods, rotifers, trochozoans, cladocerans) · multiple extant marine invertebrate clades
  • Chaetognath predator-prey ESD data used as a raptorial-predator benchmark, not entered in the regression · chaetognaths (extant)
  • Published mesh spacing and estimated prey sizes of suspension-feeding radiodonts (Aegirocassis, Pahvantia, Tamisiocaris) used for comparison · radiodonts (extinct stem-group euarthropods)
Methods & tools
Adobe Illustrator 28.2 (2024), RStudio, lm4, lmerTest (Satterthwaite approximation), performance, MuMIn, lmtest (Harvey-Collier test), stats (OLS linear regression), dplyr / plyr / magrittr, ggplot2 / ggrepel, Linear mixed-effects model with genus random effect, Cylinder volume to ESD conversion
Reagents & environment
Measurements were taken in Adobe Illustrator 28.2 (2024). Analyses ran in RStudio, with lm4, lmerTest, MuMIn and performance for mixed models, lmtest for diagnostics, ggplot2 with ggrepel for plots, and dplyr, plyr and magrittr for data handling. R and package versions are not given.
Reproducibility gaps
No code or data repository URL is given; measurements come from published images and original descriptions, with values tabulated in Table 1 and a referenced S1_OLS supplementary output.
Limitations (6)
  • The dataset is small: several mesh spacing values per species but only one trunk length each. The authors handle this conservatively with a genus random effect and a regression on aggregated means. · stated by the authors
  • Few complete specimens and likely diagenetic distortion meant body size was taken as the maximum trunk length and width from original descriptions, not necessarily from the specimen used for mesh measurements. · stated by the authors
  • The Emu Bay Shale luolishaniid is known from one incomplete specimen. Its body length (about 87.5 mm) was extrapolated from the power law, and it was left out of the main analyses to limit error. · stated by the authors
  • No luolishaniid fossil preserves gut contents or signs of food selectivity, so diet composition cannot be assessed directly; mesh size serves only as a proxy for prey. · stated by the authors
  • Luolishaniid prey ESD is not observed. It is derived from mesh spacing using extant suspension-feeder relationships, so the combined predator-prey fit (mostly driven by modern taxa) partly assumes the feeding mode it is used to support. · noted in this summary
  • Raptorial feeding is rejected against a single raptorial benchmark clade (chaetognaths), compared descriptively rather than statistically. The sediment-sieving alternative raised in the Introduction is not tested. · noted in this summary

Read this if you study Cambrian paleoecology or lobopodian functional morphology, or want a template for testing fossil feeding modes with mesh-spacing allometry and predator-prey size scaling.

Citations
1
FWCI
3.91
Field percentile
top 10%

Reproducibility ledger

Nobody has reported trying to reproduce this paper yet. Did you replicate an experiment or run the code? Your report saves the next lab a week.

Sign in to file a reproducibility report.

Related papers

Similarity Map
SelectedSame fieldCross-field
THISOrtega‐Hern '25Dassow '21Du '26Dearden '23Russell '26Kondo '18
71% semantic matchPaleontology
Javier Ortega‐Hernández·bioRxiv·Dec 30, 2025