Pith. sign in

REVIEW 3 major objections 5 minor 41 references

Nutrient Transport in Concentration Gradients

T0 review · 3 major / 5 minor · reviewed 2026-08-11 · deepseek-v4-flash

Pith's one-line read In a nutrient gradient, the treadmill ciliary stroke delivers the largest and fastest-growing nutrient intake of the motions tested, adding an amount that scales as the square root of time times Péclet number.

desk verdict Solid extension of ciliate feeding to concentration gradients with a clean analytical baseline and a clean symmetry result, but the treadmill-optimality claim is proven only for the first four modes and the scaling law is fitted, not derived. read the letter →

arxiv 2412.16408 v1 pith:C7PKBVI6 submitted 2024-12-21 physics.flu-dyn physics.bio-ph

classification physics.flu-dynphysics.bio-ph
keywords sessileciliatesciliaryfeedingcurrentssphericalenvelopemodelconcentrationgradientsunsteadyadvection-diffusionPécletnumberpatchynutrientfieldstreadmillmode
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper asks how a sessile ciliate should beat its cilia to feed when nutrients are not spread uniformly but arrive as a gradient or a patch. Using the spherical envelope model of ciliary flows and the unsteady advection–diffusion equation, it shows that in a linear concentration gradient aligned with the beat axis, only "odd" surface motions—most powerfully the treadmill stroke (mode 1)—add any nutrient intake beyond what diffusion in a uniform field provides. The extra intake grows like $0.94\,\epsilon\sqrt{t\,\mathrm{Pe}}$, so stronger flows and longer exposure both help, but sublinearly. In patchy fields, no single beat is best at all times: the optimal stroke switches depending on whether the patch is a depleted hole or a nutrient-rich blob. The result matters because it says that for attached filter feeders, directional beating and orientation relative to the gradient are as important as stroke amplitude.

What carries the argument

The machinery is Blake's spherical envelope (squirmer) model: the ciliated surface is a sphere whose tangential surface velocity is expanded in Legendre modes, $u|_{r=a} = \sum_n B_n V_n(\mu)\mathbf{e}_\theta$, with the hydrodynamic power held fixed across modes. Coupled to this is a spectral decomposition of the advection–diffusion equation into a homogeneous part $c_h$ (uniform background) and a gradient-forced part $\tilde{c}$ driven by the $z$-component of the flow, $u_z$; the symmetry properties of the Legendre modes then decide which modes contribute to the surface flux integral.

What would settle it

Recompute the same unsteady advection–diffusion problem for a surface velocity that includes modes $n \ge 5$ or a superposition of modes 1 and 3 at the same fixed power: if any such stroke yields a sustained gradient-driven uptake larger than mode 1's at, say, $\mathrm{Pe}=100$ and $t=200$, the optimality claim as stated fails. Equivalently, a microfluidic experiment with a sessile ciliate or a robotic squirmer in a known linear gradient could measure uptake for different beating patterns; observing a fore-aft symmetric stroke that still adds gradient-driven flux would contradict the even-mode-zero result.

Watch

Extended reading notes

Core claim

On its own terms, the paper establishes that in a linear concentration gradient $C_b = C_0(1 + \epsilon z)$ around a sessile spherical cell, ciliary surface motions that are fore-aft symmetric (even Legendre modes) produce zero gradient-driven nutrient uptake at all times, while odd modes produce a positive contribution. The treadmill mode $n=1$—the same stroke already known to be optimal in uniform fields—gives the largest sustained uptake, with the gradient-driven part accurately approximated by $0.94\,\epsilon\sqrt{t\,\mathrm{Pe}}$ beyond a short transient. The paper also derives an exact analytic solution for pure diffusion in a gradient showing the gradient term integrates to zero, so without flow a gradient is no better than a uniform field of the same mean concentration.

Load-bearing premise

The strongest conclusion rests on comparing only the first four Legendre modes of surface motion—and, for gradients, only the case where the beat axis is perfectly aligned with the gradient—so a stroke outside that family or a misaligned cell could, in principle, beat the treadmill mode.

Editorial extensions

If this is right

  • In an infinite linear gradient, a sessile ciliate can increase its nutrient intake without limit by persisting in the treadmill stroke, but the growth is only $O(\sqrt{t})$: the marginal benefit of staying put diminishes with time.
  • Even-mode surface motions (dipolar, quadrupolar) gain nothing from a background gradient when aligned along the symmetry axis, so any observed gradient benefit must come from an odd, directional component of the beat.
  • In patchy food fields the best stroke changes with time and patch geometry, so a ciliate that can switch beat modes—or reorient—can outperform any single fixed stroke.
  • The analytic result that pure diffusion in a gradient gives no uptake advantage isolates the role of flow: gradients are only exploitable through directional advection.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • By the same symmetry logic, any fore-aft symmetric surface velocity—not only even Legendre modes—should give zero gradient-driven uptake; this is a testable generalisation beyond the four modes the paper studies.
  • The $\sqrt{t\,\mathrm{Pe}}$ scaling suggests a natural residence-time problem: in a finite patch, the optimal time to stay before reorienting or switching modes balances the growing gradient intake against the cost of leaving other patches unexplored.
  • The results imply that for sessile ciliates attached to substrates, the angle of attachment relative to the nutrient gradient is a control variable nearly as important as the beat pattern, tying the model to observations of Vorticella orienting near boundaries.
  • The same decomposition could be applied to motile ciliates: a swimming cell's own directional motion plays the role of the treadmill mode, giving a mechanistic route from gradient sensing to chemotactic feeding.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 5 minor

Summary. The manuscript studies nutrient transport to a sessile ciliated cell modeled as a sphere with ciliary surface activity, in the presence of a linear concentration gradient. It first derives an analytical solution for pure diffusion in a linear gradient and shows that the surface-integrated uptake equals that of a uniform field. It then couples cilia-driven Stokes flow with unsteady advection-diffusion, decomposes the concentration into a uniform-background part and a gradient-driven part, and analyzes the first four Legendre modes of surface velocity at fixed dissipation power. The central quantitative finding is a fitted scaling law for mode 1, J(t) = Jh(t) + 0.94 epsilon sqrt(t Pe), with validation against numerical simulations for Pe up to 500. The paper concludes that the treadmill (mode 1) motion is optimal for sustained nutrient intake in linear gradients, while even modes give no gradient-driven contribution, and it presents two patchy-nutrient case studies comparing modes 1 and 2.

Significance. If the central claim were established for the full space of axisymmetric ciliary motions, this work would provide a useful design principle for sessile-ciliate feeding in heterogeneous nutrient fields and a clean analytical benchmark for unsteady diffusion around a sphere. The analytical pure-diffusion solution and the symmetry argument eliminating even modes are careful and convincing, and the numerical validation to 1e-4 relative error is a strength. The scaling law, while empirical, is a falsifiable prediction that could guide experiments. However, the optimality conclusion is currently demonstrated only within a four-mode family, which limits the significance of the headline claim.

major comments (3)
  1. [Abstract, Section 3 (Eq. 8), Section 4] The abstract and Section 4 state that the treadmill ciliary motion (mode 1) achieves the 'highest uptake' and that 'only the treadmill surface activity contributes significantly to sustained nutrient intake.' The analysis, however, only compares the first four Legendre modes (n = 1,2,3,4) in Eq. (8) and Fig. 2. The symmetry argument in Section 3 eliminates even modes, but it does not bound odd modes n >= 3, nor does it address superpositions of modes. Mode 3's contribution is shown to saturate, which suggests but does not prove that no higher odd mode or mixed stroke can outperform mode 1 at finite times. The optimality claim should be explicitly qualified to the family of the first four modes, or additional evidence (e.g., numerical tests of higher odd modes or a scaling argument) must be provided to support the unqualified claim.
  2. [Section 3, Eq. (17), Fig. 3] The central scaling law J(t) = Jh(t) + 0.94 epsilon sqrt(t Pe) is obtained by fitting numerical data: the coefficient 0.94 and the exponent 0.55 (approximated to 0.5) are fit parameters, as stated in the text ('We fitted the component... using a power law model'). The Summary, however, calls this an 'analytic expression.' This is a fitted empirical law, not a derived one. Additionally, the validation in Fig. 3B uses the same numerical solver that produced the fit, so it does not constitute an independent test. Please revise the wording to clearly distinguish the fitted scaling from an analytical derivation, and temper the claim that the model 'predicts' outside the fitted range.
  3. [Section 3, 'Unsteady transport in patchy concentration fields' (Figs. 4, 5)] The patchy-environment conclusions are also based on a comparison of only modes 1 and 2. The statement in Section 4 that 'shifting between ciliary beating modes optimizes cumulative nutrient intake' is therefore not an optimization over ciliary motions but rather a comparison of two modes. Please qualify these conclusions accordingly or expand the search over modes and superpositions before claiming optimality.
minor comments (5)
  1. [Title and author line] The title contains a typo: 'T ransport' should be 'Transport.'
  2. [Section 2, after Eq. (6)] There is a typo 'state-state' that should read 'steady-state.'
  3. [Section 2, Eq. (5)] The analytical solution in Eq. (5) is written with a different form in Appendix A, Eq. (42); please check consistency between the two expressions or add a note that they are equivalent.
  4. [Fig. 2 caption and panels E-G] The figure labels are somewhat confusing: panel E is referenced for uniform uptake, panel G for gradient-driven uptake, and the panel order in the caption is not fully aligned with the text. Please number panels consistently and ensure the caption describes each panel correctly.
  5. [Appendix A, Eq. (30)] The residual calculation assumes a simple pole at s = 0, but the integrand has branch points; the text should briefly justify the use of the residue theorem despite the branch cut, or cite a standard reference where this is done.

Circularity Check

1 steps flagged · score 6.0 of 10

Eq. (17)'s scaling law is a calibrated fit to the same numerical solver, then presented as an analytic prediction; the rest of the derivation is self-contained but the optimality claim is scope-limited to four modes.

  1. fitted input called prediction [Section 3, 'Unsteady transport in linear gradients', Eq. (17) and Fig. 3]
    "We fitted the component J~ of the nutrient intake using a power law model J~ = a(t)Pe^{b(t)}. ... J~(t) = 0.94(t Pe)^{0.55} ... J(t) = Jh(t) + 0.94ε√(tPe). (17) In Fig. 3B, we tested the ability of this power law to predict nutrient uptake beyond the ranges ... that were used to obtain the fitted coefficient."

    The constant 0.94 and the exponent 0.55 (approximated to 0.5) are obtained by least-squares fitting to the spectral numerical solution of Eqs. (14)-(15). Equation (17) is therefore the fitted curve rewritten, not a result derived from the transport equations. The 'prediction' check in Fig. 3B compares this fitted formula with the same numerical solver that produced the fit (Pe=100 lies inside the fitted range), so the agreement is a consistency check of the fit, not independent confirmation. Calling the fitted law an 'analytic expression' in the Summary renames a calibrated empirical power law as a derived prediction.

full rationale

The pure-diffusion section is self-contained: Eq. (5) is obtained by Laplace transform and residue calculus, and the no-advantage result follows from Legendre orthogonality (∫ Pn dµ = 2δn0). The even-mode zero result follows from front-back symmetry of Eq. (15), independent of fitting. The uniform-field ranking among modes 1-4 is recomputed numerically (Fig. 2E), so self-citations [23,25,26] are methodological rather than load-bearing. The statement that treadmill motion is optimal is, however, established only within the first four Legendre modes; this is a scope limitation, not a circular step. The one circular element is the fitted scaling law in Eq. (17) being presented as an analytic prediction. Because the central optimality comparison retains independent numerical content, the circularity is partial, hence score 6.

Assumptions & free parameters 4 free parameters · 6 assumptions · 0 invented entities

The central result relies on the Stokes flow assumption, the axisymmetric-aligned beating assumption, and a truncation to four Legendre modes. The most consequential free parameters are the fitted coefficient and exponent in the central scaling law. No new physical entities are introduced.

free parameters (4)
  • prefactor 0.94 in J = Jh + 0.94 eps sqrt(t Pe) = 0.94
    Fitted from numerical power-law fits of tilde J versus Pe and time (Fig. 3A); the R-squared is >0.99 and the value is used directly in the central formula Eq. (17).
  • fitted exponent b(t) on Pe = 0.5 (approximated from fitted b ~ 0.54)
    b(t) is fitted at each time from numerical data, then approximated to 0.5 for the final scaling law. The choice of 0.5 is a modeling decision, not a derived exponent.
  • time-growth exponent of tilde J = 0.5
    a(t) is reported to scale as t^0.5 in Fig. 3A; this is used together with the Pe exponent to form the combined sqrt(t Pe) law.
  • R(t) scaling exponents for modes 2-4 = 1, 0.5, 0.3, 0.1
    The transport-distance growth rates R ~ t, sqrt(t), t^0.3, t^0.1 are read off numerical R(t) curves (Fig. 2H). These support the mode ranking but are fitted, not derived.
assumptions (6)
  • domain assumption The fluid is governed by incompressible Stokes equations at zero Reynolds number.
    Used in Eq. (7) to model cilia-driven feeding currents. This is standard for micron-scale ciliates but is an idealized representation, not a measured flow.
  • domain assumption Ciliary surface motion is axisymmetric and aligned with the nutrient gradient.
    Boundary condition in Eq. (8) uses only theta-dependent surface velocities, and the paper explicitly states it considers surface activities axisymmetric about the concentration-gradient direction. The central conclusion depends on this alignment.
  • ad hoc to paper Only the first four Legendre modes of surface velocity are considered.
    Section 3: 'We consider surface activity associated with the first four Legendre modes.' The optimality of mode 1 is asserted within this truncated family, and patchy studies use only modes 1 and 2.
  • domain assumption The far-field boundary condition grad c -> eps ez is preserved within the considered time.
    Stated in Section 3 before Eq. (11): 'we consider far-field boundary condition preserved within the considered time.' This idealizes the infinite linear gradient as an inexhaustible reservoir.
  • domain assumption The background gradient length scale L is much larger than the cell radius a.
    Stated in Section 2: 'taken such that the length scale L >> a is larger than the cell radius a.' This justifies the small parameter eps = a/L and the perturbative decomposition.
  • standard math Linearity of the advection-diffusion equation allows the decomposition c' = ch + eps tilde c.
    Eqs. (13)-(15) split the concentration disturbance into a uniform-background part and a gradient-correction part. This is exact given the linear PDE and linear boundary conditions, but it is a structural assumption that shapes all subsequent results.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Nutrient Transport in Concentration Gradients." pith.science (2026). https://pith.science/paper/C7PKBVI6

@misc{pith2026241216408,
  author       = {Pith},
  title        = {Pith review of: Nutrient Transport in Concentration Gradients},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/C7PKBVI6}},
  note         = {Machine review of arXiv:2412.16408}
}
read the original abstract

Sessile ciliates attach to substrates and generate feeding currents to capture passing particulates and dissolved nutrients. Optimal ciliary activity that maximizes nutrient flux at the cell surface while minimizing the rate of hydrodynamic energy dissipation is well characterized in uniform nutrient fields. However, it is unclear how ciliary motion should change when nutrients are non-uniform or patchy. To address this question, we modeled the sessile ciliate and feeding currents using the spherical envelope model, and used an unsteady advection-diffusion equation to describe the nutrient scalar field. In the absence of flows, we calculated the diffusive nutrient uptake analytically in linear nutrient gradients and found no advantage over uptake in uniform fields. With ciliary activity driving feeding currents, we used a spectral method to solve for the unsteady nutrient concentration. We found that, when the axis of symmetry of the ciliary motion is aligned with the nutrient gradient, nutrient uptake at the cell surface increases steadily over time, with highest uptake achieved by the treadmill ciliary motion which is optimal in uniform fields as well. The associated nutrient uptake in concentration gradients scales with the square root of the product of time and P\'eclet number. In patchy environments, optimal ciliary activity depends on the nature of the patchiness. Our findings highlight strategies that enable sessile ciliates to thrive in environments with fluctuating nutrient availability.

Figures

Figures reproduced from arXiv: 2412.16408 by the authors.

Figure 1
Figure 1. Pure diffusion in concentration gradient. A. Mathematical model of sessile ciliated cell as a sphere of radius a, with spherical (r, θ, φ) and Cartesian (x, y, z) co￾ordinates. B. Concentration field of dissolved nutrients following a diffusion process in an initially (top row) uniform concentration (ǫ = 0) and (bottom row) gradient (ǫ = 0.05). Snapshots shown at time instances t = [0, 5, 100] and at steady-state. C… view at source ↗
Figure 2
Figure 2. Unsteady transport in uniform and linear concentration. A. Flow stream￾lines around a sessile ciliated cell generated by the first four surface velocity modes. Concentration field c in B. uniform and D. linear gradient, where c = ch +ǫc˜+ (1+ǫz). C. snapshots of c˜ with insets showing |∇c˜| around the cell. E. Time evolution of nutrient intake Jh(t) in uniform concentration. In uniform background concentration, E. m… view at source ↗
Figure 3
Figure 3. Nutrient intake of the first mode. A. Coefficients of curve fitting of nutrient intake J˜(t) corresponding to mode 1 versus Pe at every instant time. B. Nutrient intake comparison between numerical computation J(t) = Jh(t) + ǫJ˜(t) and estimation J(t) = Jh|t→∞ + 0.94ǫ √ t Pe under Pe = [100, 200, 500] with time range t ∈ [0, 500]. C. Zoom-in version for nutrient intake comparison for time range t ∈ [0, 10]. All calc… view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Nutrient uptake in a concentration field with depleted upstream patch. A. The initial depleted patchy concentration upstream of the model cell. B-C. The concentration field snapshots at time t = [0, 2, 10, 20] corresponding to surface velocity with B. mode 1 and C. mod…
Figure 5
Figure 5. Figure 5: Nutrient uptake in a depleted environment with a rich concentration patch. A. The initial field contains a concentrated blob upstream of the model cell. B-C. The concentration snapshots at time t = [0, 2, 10, 20] corresponding to surface velocity with B. mode 1 and C. …

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

41 extracted references · 41 canonical work pages

  1. [1]

    & Blackburn, N

    Fenchel, T . & Blackburn, N. Motile chemosensory behaviou r of phagotrophic pro- tists: mechanisms for and efficiency in congregating at food patches. Protist 150, 325–336 (1999)

  2. [2]

    Long, R. A. & Azam, F . Microscale patchiness of bacteriopl ankton assemblage richness in seawater. Aquatic Microbial Ecology 26, 103–113 (2001)

  3. [3]

    Durham, W. M. et al. T urbulence drives microscale patches of motile phytoplank - ton. Nature communications 4, 2148 (2013)

  4. [4]

    M., Carrara, F

    Keegstra, J. M., Carrara, F . & Stocker, R. The ecological r oles of bacterial chemo- taxis. Nature Reviews Microbiology 20, 491–504 (2022)

  5. [5]

    Marine microbes see a sea of gradients

    Stocker, R. Marine microbes see a sea of gradients. science 338, 628–633 (2012). 18

  6. [6]

    & Łagód, G

    Babko, R., Kuzmina, T ., Danko, Y ., Szul ˙zyk-Cieplak, J. & Łagód, G. Oxygen gra- dients and structure of the ciliate assemblages in floodplai n lake. Water 12, 2084 (2020)

  7. [7]

    Stal, L. J. & Cretoiu, M. S. The marine microbiome (Springer, 2022)

  8. [8]

    E., Raina, J.-B., Peaudecerf, F

    Clerc, E. E., Raina, J.-B., Peaudecerf, F . J., Seymour, J. R. & Stocker, R. in The Marine Microbiome 47–102 (Springer, 2022)

Show all 41 references
  1. [9]

    & Se ymour, J

    Raina, J.-B., Fernandez, V ., Lambert, B., Stocker, R. & Se ymour, J. R. The role of microbial motility and chemotaxis in symbiosis. Nature Reviews Microbiology 17, 284–294 (2019)

  2. [10]

    R., Johansson, M

    Jonsson, P . R., Johansson, M. & Pierce, R. W. Attachment t o suspended particles may improve foraging and reduce predation risk for tintinni d ciliates. Limnology and oceanograpghy 49, 1907–1914 (2004)

  3. [11]

    Christensen-Dalsgaard, K. K. & Fenchel, T . Increased fil tration efficiency of at- tached compared to free-swimming flagellates. Aquatic microbial ecology 33, 77– 86 (2003)

  4. [12]

    E., Roper, M., Ryu, S., Matsudaira, P

    Pepper, R. E., Roper, M., Ryu, S., Matsudaira, P . & Stone, H. A. Nearby bound- aries create eddies near microscopic filter feeders. Journal of The Royal Society Interface 7, 851–862 (2010)

  5. [13]

    Pepper, R. E. et al. A new angle on microscopic suspension feeders near bound- aries. Biophysical journal 105, 1796–1804 (2013)

  6. [14]

    Pepper, R. E. et al. The effect of external flow on the feeding currents of sessile microorganisms. Journal of the Royal Society Interface 18, 20200953 (2021)

  7. [15]

    & Prakash, M

    Krishnamurthy , D., Pepper, R. & Prakash, M. Active sinki ng particles: sessile sus- pension feeders significantly alter the flow and transport to sinking aggregates. Journal of the Royal Society Interface 20, 20220537 (2023). 19

  8. [16]

    A., Lopes, R

    Kanso, E. A., Lopes, R. M., Strickler, J. R., Dabiri, J. O. & Costello, J. H. T eam- work in the viscous oceanic microscale. Proceedings of the National Academy of Sciences 118, e2018193118 (2021)

  9. [17]

    Slabodnick, M. M. et al. The macronuclear genome of Stentor coeruleus reveals tiny introns in a giant cell. Current Biology 27, 569–575 (2017)

  10. [18]

    & Nishi gami, Y

    Echigoya, S., Sato, K., Kishida, O., Nakagaki, T . & Nishi gami, Y . Switching of behavioral modes and their modulation by a geometrical cue i n the ciliate Stentor coeruleus. Frontiers in Cell and Developmental Biology 10, 1021469 (2022)

  11. [19]

    Shekhar, S. et al. Cooperative hydrodynamics accompany multicellular-like colo- nial organization in the unicellular ciliate Stentor. bioRxiv (2023)

  12. [20]

    Nutrient fluxes to planktonic osmotrophs in the presence of fluid motion

    Karp-Boss, L., Boss, E., Jumars, P ., et al. Nutrient fluxes to planktonic osmotrophs in the presence of fluid motion. Oceanography and marine biology 34, 71–108 (1996)

  13. [21]

    A., Ganguly , S., Kessler, J

    Solari, C. A., Ganguly , S., Kessler, J. O., Michod, R. E. & Goldstein, R. E. Multi- cellularity and the functional interdependence of motilit y and molecular transport. Proceedings of the National Academy of Sciences 103, 1353–1358 (2006)

  14. [22]

    Short, M. B. et al. Flows driven by flagella of multicellular organisms enhance long-range molecular transport. Proceedings of the National Academy of Sci- ences 103, 8315–8319 (2006)

  15. [23]

    & Lauga, E

    Michelin, S. & Lauga, E. Optimal feeding is optimal swimm ing for all Péclet num- bers. Physics of Fluids 23, 101901 (2011)

  16. [24]

    & Lauga, E

    Michelin, S. & Lauga, E. Unsteady feeding and optimal str okes of model ciliates. Journal of Fluid Mechanics 715, 1–31 (2013)

  17. [25]

    Liu, J., Man, Y ., Costello, J. H. & Kanso, E. Feeding Rates in Sessile versus Motile Ciliates are Hydrodynamically Equivalent. eLife 13 (2024). 20

  18. [26]

    Liu, J., Man, Y ., Costello, J. H. & Kanso, E. Optimal Feedi ng of Swimming and Attached Ciliates. arXiv preprint arXiv:2404.13467 (2024)

  19. [27]

    Blake, J. R. A spherical envelope approach to ciliary pro pulsion. Journal of Fluid Mechanics 46, 199–208 (1971)

  20. [28]

    & Pedley , T

    Magar, V ., Goto, T . & Pedley , T . J. Nutrient Uptake by a Sel f-Propelled Steady Squirmer. The Quarterly Journal of Mechanics and Applied Mathematics 56, 65– 91 (2003)

  21. [29]

    & Pedley , T

    Magar, V . & Pedley , T . J. Average nutrient uptake by a self -propelled unsteady squirmer. Journal of fluid mechanics 539, 93–112 (2005)

  22. [30]

    Berg, H. C. & Purcell, E. M. Physics of chemoreception. Biophysical journal 20, 193–219 (1977)

  23. [31]

    Carslaw, H. S. & Jaeger, J. C. Conduction of heat in solids tech. rep. (Clarendon Press, 1959)

  24. [32]

    & Karrila, S

    Kim, S. & Karrila, S. Microhydrodynamics Butterworth 1991

  25. [33]

    Purcell, E. M. Life at low Reynolds number. American journal of physics 45, 3–11 (1977)

  26. [34]

    Goldstein, R. E. Green algae as model organisms for biolo gical fluid dynamics. Annual review of fluid mechanics 47, 343–375 (2015)

  27. [35]

    Kirkegaard, J. B. & Goldstein, R. E. The role of tumbling f requency and persis- tence in optimal run-and-tumble chemotaxis. IMA Journal of Applied Mathematics 83, 700–719 (2018)

  28. [36]

    Nelson, G. et al. Cells collectively migrate during ammonium chemotaxis in Chlamy- domonas reinhardtii. Scientific Reports 13, 10781 (2023). 21

  29. [37]

    & Zhu, L

    Zhu, G., Fang, W.-Z. & Zhu, L. Optimizing low-Reynolds-n umber predation via optimal control and reinforcement learning. Journal of Fluid Mechanics 944, A3 (2022)

  30. [38]

    Fang, W.-Z., Xiong, T ., Pak, O. S. & Zhu, L. Data-driven in telligent manipulation of particles in microfluidics. Advanced Science 10, 2205382 (2023)

  31. [39]

    & Lewis, G

    Dopheide, A., Lear, G., Stott, R. & Lewis, G. Relative div ersity and community structure of ciliates in stream biofilms according to molecu lar and microscopy methods. Applied and Environmental Microbiology 75, 5261–5272 (2009)

  32. [40]

    Sherr, E. B. & Sherr, B. F . Significance of predation by pro tists in aquatic microbial food webs. Antonie Van Leeuwenhoek 81, 293–308 (2002)

  33. [41]

    & Stegun, I

    Abramowitz, M. & Stegun, I. A. Handbook of mathematical functions with formu- las, graphs, and mathematical tables (US Government printing office, 1968). 22 APPENDIX A Diffusion in concentration gradients We consider the concentration field c′ = c − (1 + ǫz). Substitute back int...

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.