Mechanisms of temporary adhesion in benthic animals

Biol. Rev. (2011), 86, pp. 15–32. 15doi: 10.1111/j.1469-185X.2010.00132.x

Mechanisms of temporary adhesion in benthicanimals

D. Dodou1∗, P. Breedveld1, J. C. F. de Winter1, J. Dankelman1 and J. L. van Leeuwen2

1 Department of BioMechanical Engineering, Faculty of Mechanical, Maritime and Materials Engineering, Delft University of Technology,

The Netherlands2 Experimental Zoology Group, Wageningen Institute of Animal Sciences, Wageningen University, The Netherlands

(Received 30 June 2009; revised 25 January 2010; accepted 28 January 2010)

ABSTRACT

Adhesive systems are ubiquitous in benthic animals and play a key role in diverse functions such as locomotion,food capture, mating, burrow building, and defence. For benthic animals that release adhesives, surface and materialproperties and external morphology have received little attention compared to the biochemical content of the adhesives.We address temporary adhesion of benthic animals from the following three structural levels: (a) the biochemical contentof the adhesive secretions, (b) the micro- and mesoscopic surface geometry and material properties of the adhesiveorgans, and (c) the macroscopic external morphology of the adhesive organs. We show that temporary adhesion ofbenthic animals is affected by three structural levels: the adhesive secretions provide binding to the substratum at amolecular scale, whereas surface geometry and external morphology increase the contact area with the irregular andunpredictable profile of the substratum from micro- to macroscales. The biochemical content of the adhesive secretionsdiffers between abiotic and biotic substrata. The biochemistry of the adhesives suitable for biotic substrata differentiatesfurther according to whether adhesion must be activated quickly (e.g. as a defensive mechanism) or more slowly (e.g.during adhesion of parasites). De-adhesion is controlled by additional secretions, enzymes, or mechanically. Due todeformability, the adhesive organs achieve intimate contact by adapting their surface profile to the roughness of thesubstratum. Surface projections, namely cilia, cuticular villi, papillae, and papulae increase the contact area or penetratethrough the secreted adhesive to provide direct contact with the substratum. We expect that the same three structurallevels investigated here will also affect the performance of artificial adhesive systems.

Key words: benthic animals, temporary adhesion, adhesive secretions, biochemical content, surface geometry, externalmorphology.

CONTENTS

I. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16II. Biochemical content . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

(1) Adhesion to abiotic substrata . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17(2) Adhesion to biotic substrata . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

(a) Slow adhesion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19(b) Quick adhesion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

(3) De-adhesion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21(a) Duo-gland systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21(b) Mono-gland systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

(4) Diversity and convergence in the biochemical content of adhesive systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21III. Micro- and mesoscopic surface geometry and material properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

(1) Material deformability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21(2) Surface-projecting structures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22

* Address for correspondence: Tel: +31 15 278 4221; E-mail: [email protected]

Biological Reviews 86 (2011) 15–32 © 2010 The Authors. Biological Reviews © 2010 Cambridge Philosophical Society

16 D. Dodou and others

(a) Cilia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22(b) Cuticular villi . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22(c) Papillae . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22(d) Papulae . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

IV. Macroscopic external morphology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23(1) Low-curvature morphologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

(a) Extended low-curvature morphologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23(b) Localised low-curvature morphologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24

(2) Projecting morphologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24(a) Multiple-feet projecting morphologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24

( i ) Tube feet . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24( ii ) Podia in sponges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25( iii ) Arms for free walking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25( iv ) Threads for restricted walking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26

(b) Branched-feet projecting morphologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26( i ) Digitate podia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26( ii ) Peltate and pelto-digitate tentacles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27( iii ) Pinnate tentacles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

(c) Tentacular projecting morphologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27( i ) Autotomising tentacles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27( ii ) Retracting tentacles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

V. Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27(1) Biological research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28(2) Comparative studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29(3) Biomimetic potential . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

VI. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30VII. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30

VIII. References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30

I. INTRODUCTION

Benthic animals employ strong temporary attachmentmechanisms for a variety of actions, such as locomotion,food capture, mating, burrow building, and defence, towithstand water currents that would otherwise sweep themaway from their working environment. Some benthic animalsuse clamps, spines, or suction to achieve grip, whereasothers secrete adhesives. The adhesives of benthic animalscan be permanent, transitory, or temporary (Flammang,1996; Scherge & Gorb, 2001; Tyler, 1988). Permanentadhesives are secreted as fluids, which then solidify toform a cement (as in barnacles). Transitory adhesives allowmovement while sticking: The animal interposes an adhesivefilm between its body and the substratum and creeps alongit (as in gastropods). Temporary adhesives, which are thefocus herein, allow repetitive stick-unstick cycles (as inechinoderms). In practice, the distinction between transitoryand temporary adhesion is not clear-cut (Flammang, 2006)and the operating time scales of these two modes of adhesionare overlapping (Gorb, 2008).

In terrestrial environments, spiders and geckos are ableto cling to surfaces very strongly without claws andwithout releasing adhesives. These animals possess hairyattachment systems and their grip relies entirely on the micro-and mesoscopic surface geometry (relief) and macroscopicexternal morphology (shape) of their exoskeleton or footpads(e.g. Arzt, Gorb & Spolenak, 2003; Peressadko & Gorb,

2004). Dry adhesion is the effect of van der Waals forcesbetween the footpad or exoskeleton of the animal and thesubstratum. Theories are being developed that attempt todescribe the adhesion of terrestrial animals as a functionof either the surface geometry and external morphology oftheir attachment systems (Arzt et al., 2003; Peressadko &Gorb, 2004) or their phylogenetic characteristics (Peattie &Full, 2007).

For benthic animals that release temporary adhesives,the roles of surface geometry and external morphologyin adhesion have received little attention compared tothe biochemical content of the adhesives; attempts todevelop a theoretical framework describing the mechanismsof adhesion of these animals remain limited. A recentstudy by Santos et al. (2005a) showed that the adhesivefootpads of echinoderms deform in a viscoelastic manner tomatch the roughness of the substratum. Surface-projectingstructures such as cilia and papillae have been studiedextensively, but predominantly in relation to the secretoryand sensory cells present underneath the epidermis. Somehave related the external morphology of the adhesive organsin sea stars (Asteroidea) to habitat demands (Blake, 1989;Flammang, 2006; Santos et al., 2005b), but others haveargued that a consistent relationship exists between theexternal morphology of these organs and taxonomic position(Vickery & McClintock, 2000).

We review the attachment mechanisms of benthic animalsthat release temporary adhesives from three structural


Mechanisms of temporary adhesion in benthic animals 17

levels: (a) the biochemical content of the released adhesives,(b) the material deformability and the micro- (<10 μm)and mesoscopic (10–100 μm) surface geometry of theadhesive organs, and (c) the macroscopic (>100 μm)external morphology of the adhesive organs (Fig. 1). Withinthese structural levels, subcategories are defined specifyingshared mechanisms of adhesion across taxa. The functionalsignificance of these structural levels is addressed andperspectives for future biological research as well as thebiomimetic potential of biological adhesion mechanisms arediscussed.

II. BIOCHEMICAL CONTENT

The strength of the bonds between the epithelium ofthe adhesive organ and the substratum depends onthe biochemical nature of the adhesive secretions. Thetemporary character of the adhesion is controlled eitherby the release of de-adhesive secretions or mechanically. Wedistinguish between adhesive secretions suitable for abioticand for biotic substrata. The latter are further divided intosecretions for slow adhesion (e.g. in parasites) and for quickadhesion (e.g. during defence). Finally, two mechanismsof de-adhesion are discussed: duo-gland and mono-glandsystems.

(1) Adhesion to abiotic substrata

All echinoderms examined so far possess tube feet with twotypes of non-ciliated secretory (NCS) cells. These encloselarge heterogeneous granules and secrete an adhesive usedfor locomotion and anchoring to abiotic substrata, or forpicking up sand particles. The adhesive is present as a thinfilm between the cuticle of the tube foot and the substratum(Thomas & Hermans, 1985). The cuticle consists of aninternal fibrous layer (lower cuticle), a middle layer withgranules (upper cuticle), and an external layer with finefilaments (fuzzy coat) (Ameye et al., 2000). The fuzzy coatof echinoderms is particularly thick, approximately 800 nm,and probably consists of hydrophilic proteoglycans, whichmake it gelatinous (Ameye et al., 2000). McKenzie (1988)suggested that the fuzzy coat is glycocalyx with anti-foulingproperties. The adhesive secretions of echinoderms are amixture of proteins and carbohydrates in a ratio of around2:1 dry weight, with the protein fraction including chargedand polar amino acids and small-side-chain amino acids, andthe carbohydrates mostly in the form of acidic and sulphatedsugars. The secretions also include an inorganic fraction(possibly consisting of salts from sea water, see Smith &Morin, 2002) that represents more than 40% of the adhesivematerial (Flammang et al., 1998).

The mean normal tenacity (i.e. adhesive yield stress) ofechinoderm adhesives ranges between 59 and 290 kPa for thesea urchins Arbacia lixula and Paracentrotus lividus, respectively,as measured by Flammang, Santos & Haesaerts (2005).These values are comparable to the tenacity of marine

invertebrates that secrete permanent adhesives, such aslimpets and barnacles (230 kPa for both, as reviewed byFlammang, 2006). The co-presence of charged and polaramino acids and small-side-chain amino acids attributes tothese adhesives both high cohesiveness and high adhesivestrength (Flammang, 2006): the charged/polar amino acidsgenerate ionic and hydrogen bonds with the contact surface,whereas the small-side-chain amino acids increase theelasticity of the proteins, allowing the adhesive to sustainhigh deformations without rupture. The cohesiveness andadhesiveness of the secretions are further a trade-off betweenthe degree of crosslinking and the chain length between thecrosslinks: increasing the degree of crosslinking makes thematerial more cohesive but leaves less side chains free foradhesion, whereas with longer chains a material is moreadhesive but less cohesive. The protein fraction containsrelatively high amounts of glycine, proline, isoleucine, andcysteine. Cysteine is responsible for intermolecular disulfidebonds that increase the cohesiveness of an adhesive, andglycine is characteristic of strong, tightly bound proteinssuch as those in silk.

The strength of echinoderm adhesives is further enhancedby the fact that the released material consists of a sponge-like matrix with 3–10 μm holes defined by walls of 0.4μm thickness, which inhibits the propagation of cracksand increases its adhesive strength (Flammang et al., 1998)(Fig. 2A). The spongy appearance is most pronounced inthe thinnest areas of the footprint; when larger quantitiesof adhesive are secreted, the footprint exhibits a felt-like appearance (Flammang et al., 1998). For mussels andbarnacles it has been found that proteins are self-organisedto create the porous layer (Wiegemann & Watermann,2003). For echinoderms less is known about the exactmechanism underlying the development of their adhesivematrix. Recently, Hennebert et al. (2008) provided evidenceabout the roles of the two types of NCS cells in thedevelopment of the matrix: one type of cell produces ahomogeneous film that covers the substratum and theother type of cell releases heterogeneous electron-densematerial that expands and fuses into the homogeneous film.Hennebert et al. (2008) superimposed adhesive pores on theadhesive matrix and found a match indicating the adhesivecells possibly form a template for casting the matrix pattern.

Apart from their common general working principleand gross biochemical similarities, the adhesive secretionsof echinoderms exhibit function- and habitat-relateddifferentiation at the glandular and cellular level. Inasteroids, species confined to hard rocky substrata havecomplex NCS cell granules enclosing a highly organisedcore. Soft-substratum-dwelling species, on the other hand,have granules with a simpler ultrastructure (Engster &Brown, 1972). A mechanistic explanation of this habitat-related difference is still to be found. In sea urchins,the epidermal cells are flask-shaped in coronal podiaand cylindrical in peristomeal podia. This differentiationis possibly function-related: stronger adhesion is requiredfor coronal podia anchoring to a substratum than



Surf

ace-

proj

ectin

g st

ruct

ures

Mat

eria

l def

-or

mab

ility

Mic

ro/m

esos

copi

c su

rfac

e ge

omet

ry a

nd m

ater

ial p

rope

rtie

s

Cili

a

Mac

rosc

opic

ext

erna

l mor

phol

ogy

Papu

lae

Papi

llae

Cut

icul

arvi

lli

Low

-cur

vatu

re

mor

phol

ogie

s

Loc

alis

ed

low

-cur

vatu

rem

orph

olog

ies

Ext

ende

d lo

w-c

urva

ture

mor

phol

ogie

s

Thr

eads

for

rest

ric-

ted

wal

king

Arm

s fo

rfr

ee w

alk-

ing

Proj

ectin

g m

orph

olog

ies

Kno

b-en

ding

Dis

c-en

ding

Podi

a in

spon

ges

Tub

e fe

et Pe

nic-

illat

e

Ret

ract

ing

Aut

oto-

mis

ing

Pinn

ate

Pelta

teD

igita

te

Ten

tacu

lar

proj

ec-

ting

mor

phol

ogie

s B

ranc

hed-

feet

pr

ojec

ting

mor

phol

ogie

sM

ultip

le-f

eet

proj

ectin

g m

orph

olog

ies

Mol

ecul

ar s

cale

Bio

chem

ical

con

tent

Mic

rosc

ale

Mes

osca

leM

acro

scal

e

PRO

VID

E M

OL

EC

UL

AR

BIN

DIN

GIN

CR

EA

SE C

ON

TA

CT

AR

EA

De-

adhe

sion

Duo

-gl

and

syst

ems

Mon

o-gl

and

syst

ems

Adh

esio

n fo

r ab

iotic

subs

trat

a

Adh

esio

nfo

r bi

otic

su

bstr

ata

Slow

Qui

ck

10 μ

m10

0 μm

Adh

esio

n

Fig

.1.

The

prop

osed

stru

ctur

alle

vels

ofte

mpo

rary

adhe

sion

mec

hani

sms.

The

grey

surf

aces

repr

esen

tthe

adhe

sive

area

sin

the

line

draw

ings

.The

illus

trat

ions

are

nott

osc

ale.



A B C

D E F G

Fig. 2. (A) Adhesive footprint of the asteroid Asterias rubens (Flammang et al., 1998). (B) Adhesive footprint of the platyhelminthTroglocephalus rhinobatidis (Hamwood et al., 2002). Note the similar spongy structure of (A) and (B). (C) Cuvierian tubules of theholothuroid Holothuria forskali (lower left) immobilising a crab (upper right) (scale not reported in the original) (Flammang et al., 2002).(D, E) Duo-gland adhesive system in A. rubens. Longitudinal sections before attachment (D) and after detachment (E) CS: ciliatedsecretory cell, CU: cuticle, FC: fuzzy coat, NCS1: type 1 non-ciliated secretory cell, NCS2: type 2 non-ciliated secretory cell, P: pore,SCC: subcuticular cilium, SG: secretory granule (Flammang et al., 2005). (F, G) Anterior adhesive area in T. rhinobatidis. Longitudinalsections with everted duct endings (arrows) during attachment (F) and with retracted duct endings (arrows) after detachment (G)A: adhesive, I: integument, ES: electron-dense spheroidal secretions, RS: rod-shaped secretions (Whittington et al., 2004). The insetscorrespond to the structural level illustrations in Fig. 1.

for peristomeal podia wrapping particles (Flammang &Jangoux, 1993). A unique adhesive system has beenfound in the tentacles of the holothurian Leptosynapta sp.(Holothurioidea, Echinodermata) (McKenzie, 1988). Thetentacles of Leptosynapta sp. contain secreting goblet cellsnot found in any other dendrochirote or aspidochiroteholothurian (Bouland, Massin & Jangoux, 1982; Fankboner,1978; McKenzie, 1987), and the pH of the secreted mucus isbasic, therefore having weaker de-adhesive properties thanif it had been acidic (Hermans, 1983). Leptosynapta sp. has adelicate, worm-like apodous morphology and its tentacles areused to consolidate and lubricate its burrow. The propertiesof the adhesive system of Leptosynapta sp. are thus likelyto be related to the lifestyle of this animal, which differsfrom that of dendrochirote and aspidochirote holothurians(McKenzie, 1988).

The arms of the brachiolaria larvae of sea stars carry adhe-sive cells that contain molecules similar to those found in theadhesive system of the adult sea stars and which are suitablefor attaching to abiotic surfaces during exploration (Haesaertset al., 2005a). Other benthic animals possessing adhesive sys-tems suitable for abiotic substrata include the archiannelids

Protodrilus sp., Saccocirrus sonomacus, and Saccocirrus eroticus

(Martin, 1978), the turbellarian Monocelis cincta (Martin,1978), several gastrotricha (Tyler & Rieger, 1980), nema-toda (Adams & Tyler, 1980), and the cephalopods Euprymna

scolopes and Idiosepius sp. (Von Byern & Klepal, 2006).

(2) Adhesion to biotic substrata

(a) Slow adhesion

Whittington & Cribb (2001) introduced the term ‘‘tissueadhesion’’ to define adhesion of organisms to biotic substrata(i.e. living individuals), such as the adhesion of parasiticorganisms to the skin of their hosts. Tissue adhesivesare remarkably strong, able to bond to the current-sweptskin of the host organism. They can be permanent (as inparasitic barnacles), transitory (as in parasitic gastropods), ortemporary (as in parasitic monogeneans). Here we focus onthe latter type.

Temporary tissue adhesives have been found andcharacterised in a number of monogenean (platyhelminth)parasites, mainly Capsalidae (Hamwood et al., 2002). Theadhesive secretions of the monogenean parasite Entobdella



soleae as well as of other monogeneans such as Troglocephalus

rhinobatidis (Fig. 2B) are similar in appearance to those ofasteroids (Fig. 2A): they have similar amino acid profilesand are porous, insoluble, spongy, soft, proteinaceous, andwith networks of strands (Hamwood et al., 2002). However,compared to sea star adhesives, monogenean adhesivescontain neurophysin and keratin, higher concentrations ofglycine and threonine, lower concentrations of isoleucine,and they lack carbohydrates and lipids. The presence ofkeratin is crucial for the toughness of tissue adhesives: keratinis a tightly folded protein with helices interlinked via disulfidebonds; when stresses are applied, a domain unfolds, actingas an effective energy absorber and preventing breakageof strong bonds in the backbone proteins. Furthermore,although in both sea stars and monogeneans the adhesivesare stored in rod-shaped bodies, outer membranes of thesebodies show up only in the glue matrix of the monogeneansecretions and not in sea star secretions. Hamwood et al.(2002) suggested that these structural differences betweenmonogenean and asteroid adhesives may be critical for thestrong attachment of monogeneans to their living hosts.

A different adhesion mechanism than that in E. solea hasbeen found in the monocotylidan monogenean Neoheterocotyle

rhinobatidis and Troglocephalus rhinobatidis. Whereas in E. solea

the adhesives are secreted by two permanently exposedadhesive pads, monocotylids possess anterior apertures.When these flatworms approach a substratum, they extendanteriorly to initiate contact and open the apertures sothat the duct endings are everted and secrete an adhesive(Whittington, Armstrong & Cribb, 2004). When the animalresumes swimming, the duct endings are retracted into theapertures (see also Section II.3b).

Cyprids, the last mobile larval form of barnacles, exploresurfaces by using a pair of antennules that are able to attach,detach, and reattach to a surface, allowing the cyprid to‘‘walk’’ in a bipedal fashion. Attachment is mediated by thesecretion of a protein (Callow & Callow, 2001; Walker& Yule, 1984). Recently, Phang et al. (2010) conductedmechanical testing of the proteins in the cyprid footprintand found that the individual fibrils of the adhesive exhibitanisotropic properties: they are interconnected by thin fibres.It is possible that these fibres act as sacrificial bonds andfail first, absorbing energy, thereby preventing breakageof backbone proteins – similar to the spongy material ofechinoderm adhesives described in Section II.1.

(b) Quick adhesion

A small number of benthic animals have systems that releasequick-setting adhesives for capturing prey and immobilisepredators. Such a mechanism is used for prey capture by thetentilla of Coeloplana bannworthi (Ctenophora, Platyctenida)(Eeckhaut et al., 1997; see also Franc, 1978) and by theCuvierian tubules of sea cucumber species (Endean, 1957;Flammang, Ribesse & Jangoux, 2002).

Cuvierian tubules are stored in large numbers (200–600in Holothuria forskali) in the posterior cavity of sea cucumbers(VandenSpiegel & Jangoux, 1987; VandenSpiegel, Jangoux

& Flammang, 2000). When disturbed by a predator, a seacucumber expels 10–20 Cuvierian tubules, which, as soonas they contact the water, elongate by up to 20 times theiroriginal length (from 2.5 cm to 50 cm; VandenSpiegel &Jangoux, 1987) and stick to the predator in less than 10 s,which is considerably quicker than most biological adhesives(DeMoor et al., 2003) (Fig. 2C).

The mean normal tenacities of Cuvierian tubules varybetween 30 and 135 kPa (DeMoor et al., 2003), which istowards the low end of the range of tenacities measuredfor other temporary adhesives. Quantitative studies on theamount of adhesive released are missing, but this may belarger than that of adhesives secreted during functions suchas locomotion, considering that defence is an important butrelatively infrequent activity.

The adhesion of Cuvierian tubules is derived from theirmesothelium, which consists of two layers, an outer protectivelayer and an inner one with densely packed granular cells.When the tubules are expelled from the body and expand,the outer layer disintegrates exposing the granular cells,which expel their contents and adhere to the body of thepredator. The material stored in the cells prior to releaseis rich in protein and amino acids with low molecularweight (around 10–19 kDa). The released material includesproteins with molecular weights ranging between 10 and220 kDa. The proteins contain amino acids closely relatedto those in the stored material suggesting that the proteins inthe secreted material result from polymerisation of the lowmolecular weight protein (DeMoor et al., 2003). How thispolymerisation occurs within few seconds upon release of theprotein in water remains unknown.

The adhesive released by Cuvierian tubules is a 3:2 mix-ture of proteins and carbohydrates and contains a highproportion of insoluble proteins (DeMoor et al., 2003). Theprotein fraction contains a high percentage (more than 70%)of charged and polar amino acids, but its polarity is rela-tively low (around 45%) compared to other marine adhesives(Flammang, 2006). Still the adhesives of Cuvierian tubulesare the most hydrophilic among the marine adhesives inves-tigated so far. The combination of decreased polarity andincreased hydrophilicity in Cuvierian tubules perhaps indi-cates a trade-off between reduced adhesive strength andincreased activation speed. Although decreased polarity usu-ally relates to decreased hydrophilicity, in marine adhesivesthese two properties are not mutually antagonistic (Flam-mang, 2006), because marine adhesives consist of a mixtureof different proteins. The protein fraction in the adhesivereleased by Cuvierian tubules is rich in glycine (DeMoor et al.,2003), almost matching that in mussel adhesives, whereasthe carbohydrates are in the form of neutral sugars only, notacidic sugars as in the adhesives used for abiotic substrata.The inorganic residue is only 11% – considerably less thanthe 40% in the adhesives used for abiotic substrata. The highpercentage of carbohydrates as well as the presence of pre-cursor proteins with low molecular weight, which polymeriseupon release, may explain the rapid activation of adhesionby Cuvierian tubules (Flammang & Jangoux, 2004).



(3) De-adhesion

(a) Duo-gland systems

The duo-gland adhesive system is one of the most widespreadmechanisms of temporary adhesion in benthic animals.All echinoderm species possess such a system with twotypes of NCS cells secreting an adhesive and ciliatedsecretory (CS) cells enclosing small homogeneous electron-dense granules and controlling de-adhesion (Flammang,2006; Hermans, 1983). De-adhesion occurs at the outermostfuzzy coat of the cuticle (Flammang, 1996). According toone hypothesis, the de-adhesive function of the CS cellsis enzymatic, releasing the fuzzy coat from the underlyingcuticular layer [indeed, as Flammang (1996) reported, intransmission electron micrographs of detached podia, thefuzzy coat was no longer distinguishable] (see also Flammang,Demeulenaere & Jangoux, 1994; Flammang et al., 1998,2005) (Fig. 2D, E). After de-adhesion, the secreted adhesiveis left behind on the substratum as a footprint. Othersproposed that the role of CS cells is ‘‘neurosecretory-like’’, directly controlling or terminating the release of theadhesive (Ball & Jangoux, 1990). According to this hypothesis‘‘[considering that] the cuticle in some echinoderms mayact as an anti-fouling surface and will only function inadhesion for so long as an adhesive secretion is being activelysecreted (McKenzie 1987) . . . there is no requirement fora separate de-adhesive secretion’’ (Ball & Jangoux, 1990,p. 208).

(b) Mono-gland systems

In a number of benthic animals, although adhesion iscontrolled biochemically, de-adhesion is achieved withoutthe secretion of additional material. In the monogenean(platyhelminth) parasite E. soleae, two different materials arereleased to its host, and it was initially hypothesised thatthe two secretions are products of a duo-gland adhesivesystem (Tyler, 1988). However, Kearn & Evans-Gowing(1998) showed that both secretions are adhesive, interactingto form a glue, and that de-adhesion is most likely controlledby enzymes found at the anterior adhesive area of theintegument.

Other animals control de-adhesion mechanically: Themonocotylidan monogeneans N. rhinobatidis and T. rhinoba-

tidis possess apertures than open and close and duct endingsthat evert during adhesion and retract during detachment.In this way, the animal can pull itself away from the sub-stratum mechanically (Whittington et al., 2004) (Fig. 2F, G).Cuvierian tubules autotomise, leaving the sea cucumberfree to crawl away (see Section IV.2ci). The tentacles ofNautilus pompilius detach by bending movements of theirmusculature (Kier, 1987). Cyprids also control de-adhesionwithout the secretion of additional material, although lit-tle is known about the detachment mechanism of theselarvae.

(4) Diversity and convergence in the biochemicalcontent of adhesive systems

Flammang et al. (2005) investigated the variability of anumber of adhesives and found that, unlike permanentadhesives, temporary adhesives of disparate benthic phylashare common amino acids, indicating ‘‘convergencein composition because of common function (i.e. non-permanent attachment to the substratum) and selectivepressures’’ (p. 207). Adhesives of duo-gland systems aresimilar to tissue adhesives, with relatively high amountsof glycine, proline, isoleucine, and cysteine, whereas quickadhesives form a distinct group with particularly high levelsof glycine (254–298 residues per thousand in the adhesivesof Cuvierian tubules of four holothurians vs. 97 residuesin Asterias rubens and 124 in E. solea) (Flammang, 2006;Hamwood et al., 2002).

III. MICRO- AND MESOSCOPIC SURFACEGEOMETRY AND MATERIAL PROPERTIES

The efficacy of an attachment depends on the qualityof the geometrical match between the adhesive organand the substratum. This match is enhanced at a micro-and mesoscopic level by: (a) material deformability, theability of the adhesive organs to match the complex andunpredictable profile of the substratum, and (b) surfacestructures, namely cilia, cuticular villi, papillae, and papulae.Past studies primarily related these surface structures tothe presence of sensory and secretory cells lying in andunderneath the epidermis. Here we focus on the mechanicalcontribution of these surface structures to adhesion. Materialdeformability, cilia, and cuticular villi affect adhesion ata microscale, and papillae and papulae contribute at amesoscale.

(1) Material deformability

Santos et al. (2005a) showed that sea urchins and sea starsexhibit higher adhesion on rough than on soft substrata.They also showed that this happens because the tube feetdeform in a viscoelastic manner to match the profile of thesubstratum (Fig. 3A, B). Under slow self-imposed forces thetube foot exhibits viscous behaviour to adapt to the roughnessof the substratum, whereas under short pulses of wave-generated forces, the attached tube foot behaves elasticallyto distribute the stress homogeneously along the area ofcontact. The adhesive secretions fill only the small surfaceirregularities in the nanometre range. This mechanism ofadhesion enhancement by material deformability resemblesthe attachment mechanisms of grasshoppers (Gorb, Jiao &Scherge, 2000) and tree frogs (Scholz et al., 2009) in terrestrialenvironments, which deform their pads viscoelasticallyto maximise contact with the substratum. Santos et al.(2005a) were the first to report such behaviour in benthicanimals.



A B C

D E F

Fig. 3. (A, B) Paired images showing viscoelastic deformations of the tube foot disc of the echinoid Paracentrotus lividus (right) to theprofile of the substratum (left): poly(methyl methacrylate) (A) and polypropylene (B) (Santos et al., 2005a). (C) Tuft of vibratile ciliain the ctenophore Coeloplana bannworthi. VC: vibratile cilia (Eeckhaut et al., 1997). (D) Cuticular villi on the antennules of the cypridof the barnacle Balanus amphitrite (Phang et al., 2008). (E) Papillae on the buccal cones of the pteropod mollusc Clione limacina. Notethe projecting rosettes (arrows). c: motile cilia, p: papillae, sc: tufts of cilia (Hermans & Satterlie, 1992). (F) Papulae of the adhesiveknob in the sponge Tethya seychellensis (scale not reported in original). G: globiferous cells, P: papulae (Fishelson, 1981). The insetscorrespond to the structural level illustrations in Fig. 1.

(2) Surface-projecting structures

(a) Cilia

Cilia are fine 5–10 μm long projections of cells and canbe non-motile or motile. Non-motile cilia act as sensoryorganelles. Here we focus on motile cilia.

Motile cilia beat in a fluid (e.g. mucus) during locomotion,propulsion, or food transfer. Motile cilia have been found in anumber of larvae, such as the barrel-shaped doliolaria larvaeof crinoids. These larvae are equipped with an attachmentcomplex of a ciliary cap with secretory cells, an apical tuft withsensory cells, and an adhesive pit (Jangoux & Lahaye, 1990;Nakano et al., 2003). Each of the sensory cells bears a longcilium which is vibratile and beats in the mucus producedby the secretory cells, while the tuft brushes the substratum.The combined action of the secretory and sensory cellsallows the larva to move and explore the substratum whileremaining loosely attached to the water-substratum interface(Flammang, 1996; Jangoux & Lahaye, 1990).

In the ctenophore Coeloplana bannworthi, an indirect role ofvibrating cilia in adhesion has been observed. This animalhas two tentacles with numerous tentilla. Six different typesof cells have been found on the tentilla surface, includingcells with 7-μm long cilia, cells with short pegs, and adhesivecells (collocytes). According to Eeckhaut et al. (1997), the cilia(Fig. 3C) are longer than the pegs, and therefore are thefirst to be stimulated by vibrations caused by prey. Thisactivates the motion of the tentilla which contract and bring

the prey closer to the tentillum surface. As soon as the preytouches the pegs at the tentillum surface, the pegs stimulatethe nearby collocytes via connecting nerves, the collocyteselevate slightly, and the prey sticks to them.

(b) Cuticular villi

Cuticular villi are fine 1–2 μm long hairy structures on theadhesive organs of cyprids and contribute to adhesion in aunique way. Phang et al. (2008) observed cuticular villi at theterminals of the antennules of the Balanus amphitrite cyprid(Fig. 3D) and developed a number of hypotheses about therole of villi in cyprid adhesion. According to one of thesehypotheses, the villi penetrate the adhesive and contact thesubstratum directly in a similar manner to the spatulaeof geckos. The role of the secreted material is to displacewater to provide a local environment with a lower dielectricconstant than water, therefore increasing the interactionforces between the villi and the substratum. If this hypothesisis correct, cyprids are the only animals known to utilise acombination of wet and dry attachment underwater (Phanget al., 2008; see also Aldred & Clare, 2008).

(c) Papillae

The adhesive organs of benthic animals are generally coveredwith adhesive papillae: protruding surface structures throughwhich adhesives are released. In the buccal cones of the



mollusc Clione limacina Hermans & Satterlie (1992) observedthat, in addition to secreting adhesive material, papillae thatare 15 μm high and 10 μm in diameter also contribute toadhesion mechanically. When the buccal cones are retracted,the epidermis between clusters of papillae is folded, whereaswhen the buccal cones are extended, the clusters stand proud(Fig. 3E). These observations led Hermans & Satterlie (1992)to conclude that food-capture mechanism of Clione limacina isnot only due to chemical adhesion; papillae enhance grippingvia mechanical interlocking.

The larval tentacles of the dendrochirote holothurian Aslia

lefevrei bear numerous approximately 200 μm long projectingpapillae that contribute mechanically to adhesion by increas-ing the area of contact (Costelloe, 1988). Keogh & Keegan(2006) found that the tube feet of the ophiuroid Amphiura

filiformis are also covered with papillae, whereas the tube feetof Amphiura chiajei are smooth, and they attributed this differ-ence to the feeding styles of the two species: A. filiformis is asuspension feeder and requires a large filtering area, achievedby means of the sticky papillae. The smooth podia of A. chiajei,on the other hand, are sufficient for deposit feeding.

(d) Papulae

A peculiar surface structure has been found on thepodial surface of two sponge species, Tethya seychellensis

and T. aurantium (see Section IV.2aii, for discussion on themotile podia in sessile sponges). In a study on the transportcapabilities of these two sponges, Fishelson (1981) describedthe presence of papulae with knobby ends. Although ofsimilar shape to the papillae described above, these papulaewere larger: approximately 90–100 μm long and 60–80 μmwide (Fig. 3F). The papulae were visible before the initiationof adhesion, but flattened and disappeared when the podiacame into contact with the substratum making the surface ofthe adhesive knob smooth and possibly more adaptable tothe profile of the substratum.

IV. MACROSCOPIC EXTERNAL MORPHOLOGY

We identified two types of macroscopic external morphologyrelevant to properties of the structures that the animalneeds to adhere to: (a) low-curvature morphologies and (b)projecting morphologies. In low-curvature morphologies,adhesives are secreted along the mantle of the animal,whereas in projecting morphologies, adhesives are secretedfrom the extremities. We distinguished two groups of low-curvature morphologies (extended and localised) and threegroups of projecting morphologies (multiple feet, branchedfeet, and tentacular appendages).

(1) Low-curvature morphologies

(a) Extended low-curvature morphologies

For relatively demanding functions such as strong temporaryattachment in harsh habitats, benthic animals exhibitextended low-curvature morphologies, in which adhesionis activated and de-activated over a large area. Two cases ofsuch adhesive systems have been described in the literature:the cephalopods Sepia spp. which use a combination ofadhesion and suction, and the cephalopod Euprymna scolopes

which uses temporary adhesion for camouflage.Sepia spp. (Mollusca, Cephalopoda) live in wave- and

windswept rocky habitats with turbulent water flows andrequire quick and very strong attachment. To achieve that,at least three species of Sepia (S. tuberculata, S. typica, andS. papillata) use a combination of suckers and adhesivessecreted along the ventral surface of the mantle of theanimal and the posterior surface of the ventral arms(see Von Byern & Klepal, 2006, for a review) (Fig. 4A).Behavioural studies for S. papillata are missing, but for S.

tuberculata and S. typica the mantle first contracts to createa sucker-like cavity for quick attachment following whichan adhesive is released to reinforce the bonding with thesubstratum. In S. tuberculata the adhesive area has ridges thatpresumably enhance attachment (Roeleveld, 1972) (Fig. 4A).

1 cm

A B

Fig. 4. Low-curvature morphologies. (A) Ventral surface of the cephalopod Sepia tuberculata showing the adhesive area (black arrows)on the mantle (Von Byern & Klepal, 2006). (B) Morphology of the mantle (white arrow) and fin (black arrow) of the cephalopodIdiosepius biserialis (Von Byern & Klepal, 2006). The insets correspond to the structural level illustrations in Fig. 1.



Histochemical analyses reviewed by Von Byern & Klepal(2006) showed the presence of two kinds of glandular cellthat may both contribute to adhesion, but a comprehensiveunderstanding of this adhesive mechanism as well as themechanism that initiates suction is missing.

Euprymna scolopes uses an extended low-curvature adhesivesystem for camouflage. This cephalopod burrows under sandand cements sand grains to its body using adhesives secretedfrom its dorsal epidermis (Moynihan, 1983a; Singley, 1982).Shears (1988) observed that E. scolopes did not use a sandcoat at night but only when forced to emerge from thesand during the day for feeding. The animals stayed at amaximum distance of two body lengths from the substratum,where camouflage would be most effective. Shears (1988) alsofound that a sand coat inhibited the animal’s movements,making prey capture more difficult and suggesting that thecamouflage functions as protection from predators ratherthan in assisting prey capture. The sand coat also acts as adefensive mechanism: when threatened, the animal releasesthe sand instantaneously to distract the predator (Shears,1988). Shears (1988) further suggested that the sand coatmay serve in sediment consolidation, preventing debris fromentering the mantle and allowing the animal to breathe whenburied in the sand; the use of the sand coat for camouflagemay have evolved from this initial function. The mechanismcausing rapid de-adhesion of such a large area (the entiremantle) remains unexplored.

(b) Localised low-curvature morphologies

In flat water and stagnant environments the attachmentarea does not need to be extended; the adhesive functioncan be localised to a relatively small area of the epidermis.For example the tiny cephalopod Idiosepius sp., at just a fewmillimetres long, uses a small adhesive area in the posteriorregion of its mantle and fins to temporarily attach to seagrass or algae for camouflage (Moynihan, 1983b; Von Byernet al., 2008) (Fig. 4B). The absence of musculature and nerve

fibres connected to the glandular cells of these regions rulesout the hypothesis that the animal uses suckers. Its adhesionis probably achieved by means of a viscous mucous layer ora duo-gland system as those described in Section II.3a.

(2) Projecting morphologies

(a) Multiple-feet projecting morphologies

Multiple-feet projecting morphologies consist of proximalflexible stems endings in sensory/secretory/adhesive apices.From the adhesive systems found in the literature, wedistinguished four main designs: tube feet, podia in sponges,arms for free walking, and threads for restricted walking.

( i ) Tube feet. Tube feet occur in echinoderms as externalappendages of the water-vascular system and consist ofa proximal flexible and extensible stem ending in a distalarea that secretes adhesive material. Whereas the histologicalstructure of tube feet is similar for all echinoderms (consistingof an inner mesothelium around the water-vascular lumen,a connective tissue layer, a nerve plexus, and an outerepidermis covered with a cuticle), a number of tube-footmorphotypes exist: disc-ending, knob-ending, and penicillate(reviewed by Flammang, 1996). Flammang (1996) discussedthree other types of tube feet as well: lamellate, ramified,and digitate podia. Lamellate podia (knob-ending podiawith a flattened and folded stem) are respiratory and non-adhesive, and therefore not included here. Ramified (peltateor pinnate) and digitate podia are discussed under branched-feet projecting morphologies (Section IV.2b).

Disc-ending tube feet consist of a flexible stalk (thestem) that ends in a flattened adhesive area (the disc)(Fig. 5A). Function- and habitat-related variations in disc-ending tube feet morphology have been observed. Inasteroids living in rocky intertidal areas, the disc is reinforcedwith collagen fibres that enable stronger attachment thansimple disc-ending tube feet do (Santos et al., 2005b). Inechinoidea, disc-ending adoral tube feet usually have a

A B C

Fig. 5. Tube-foot morphotypes. All three consist of a stem ending in an adhesive disc of various forms (arrows). D: disc, S: stem.(A) Simple disc-ending tube foot of the echinoid Heterocentrotus trigonarius (Santos et al., 2009). (B) Knob-ending tube foot of the asteroidAstropecten irregularis (Santos et al., 2009). (C) Penicillate tube foot of the echinoid Echinocardium cordatum (Flammang, 1996). The insetscorrespond to the structural level illustrations in Fig. 1.



thicker, stronger, and more extensible stem, and an enlargeddisc as compared to aboral feet (Flammang & Jangoux,1993; Leddy & Johnson, 2000; Santos & Flammang, 2005;Smith, 1978). This differentiation is possibly function-related(Flammang & Jangoux, 1993): Adoral tube feet are usedin locomotion and anchorage, whereas aboral tube feetfunction in postural changes and food capture. The tube footdiameter, number, thickness, density, and development alsovary with functional requirements. Unlike the suggestion ofSmith (1978), larger animals do not necessarily possess alarger number of tube feet. Increased adhesion can resultfrom the combined effects of variables such as tube footnumber and arrangement, mechanical properties, etc. (seeSantos & Flammang, 2005). Disc-ending tube feet have alsobeen found in holothuroids (Flammang & Jangoux, 1992)and sand dollars (Clypeasteroida) (Mooi, 1986a, b; Nichols,1959; Telford & Mooi, 1986; Telford, Mooi & Harold, 1987).

Knob-ending tube feet consist of a stem ending in apointed knob (Fig. 5B). Within the asteroids, Paxillosidaare the only order possessing adhesive knob-ending tubefeet (Santos et al., 2005b; Vickery & McClintock, 2000).Paxillosida bury themselves completely within the sedimentand use their adhesive tube feet for locomotion as well as forsand burrowing. Vickery & McClintock (2000) reporteda variation on knob-ending tube feet in the paxillosidMacroptychaster accrescens which has slightly rounded tips.The exact adaptive value of knob-ending tube feet remainsunclear.

Irregular echinoids of the order Spatangoida possesspenicillate tube feet (Flammang, 1996). These are similarto simple disc-ending tube feet, but their adhesive disc hasnumerous digitations, covering either the entire disc or onlyits margin (Fig. 5C). Spatangoids live in burrows in thesediment and use their tube feet for various functions: Thoselocated close to the mouth are used for collecting particlesfrom the floor of the burrow and for transporting themto the mouth; those that are located in the posterior anddorsal regions build and maintain the burrow (Flammang,De Ridder & Jangoux, 1990; Nichols, 1959).

( ii ) Podia in sponges. Sponges are traditionally consideredas sessile metazoans. Fishelson (1981), however, observed

that individuals of two sponge species, Tethya seychellensis andT. aurantium, have podia consisting of a 10–16 mm longand 0.3–1.0 mm wide flexible stem with a swollen, adhesiveknob at the distal end (Fig. 6A). Dormant while the spongeremains undisturbed, these podia are activated when theanimal is covered by sediment, where upon they extend theirstem, sticking the knob at a distance from the sponge. Theknob becomes flattened as it adheres to the substratum andthe papulae on its surface disappear (see Section III.2d). Thepodia then shorten and pull the entire sponge along via alever action. By this method the sponge can be moved to anew site.

Although the external morphology of these podia hassimilarities with the tube feet of echinoderms (proximalstem, distal adhesive end), their internal morphology iscompletely different. Sponge podia lack the water-vascularsystem that characterises echinoderm tube feet; instead, theyhave a gelatinous core containing bundles of microfibers andmyocytes, primary archaeocytes, nucleated archaeocytes,and scleroblastic cells, all of which contribute to the mobilityof the sponge podia. The myocytes are contractile and arepartially responsible for the shortening of the podia, whereasthe nucleated archaeocytes undergo a process of ripeningto produce collagenic filaments deposited in the adhesivematrix. The gelatinous core is surrounded by a layer ofglobiferous (excretory) cells (see Fig. 3F); these are bottle-shaped with their neck reaching on the sponge surface andresponsible for releasing adhesives. Sponge podia are capableof directional movement, similar to that of sea urchin tubefeet, but at much slower time scales.

( iii ) Arms for free walking. In the last pre-metamorphicphase, during which the larvae search for a suitable siteto settle for metamorphosis, larvae have mechanisms toprevent being swept away by water currents (e.g. Butman,Grassle & Webb, 1988; Mullineaux & Butman, 1991; Pawlik,Butman & Starczak, 1991). The brachiolaria larvae of seastars are equipped with three arms and an adhesive disc(Haesaerts et al., 2005a; Haesaerts, Jangoux & Flammang,2005b). Each arm consists of a stem ending in a distaladhesive crown (Fig. 6B). The brachiolaria larva exploresthe environment, and when it detects a suitable substratum

BA C

Fig. 6. Multiple-feet projecting morphologies. (A) The sponge Tethya seychellensis with some of the extended podia free and otheradhered [scale is not given in the original, but Fishelson (1981) reported that the podia are 10–16 mm long, for sponges 2–3 cm indiameter]. AP: adhered podia, FP: free podia (Fishelson, 1981). (B) Arm of a brachiolaria larva of the sea star Asterias rubens. Thearm consists of a stem and ends in an adhesive apex. AP: adhesive apex, S: stem (Haesaerts et al., 2005b). (C) Cyprid of the barnacleBalanus amphitrite with adhesive antennules (arrow) (Phang et al., 2008). The insets correspond to the structural level illustrations inFig. 1.



for metamorphosis, it directs two or three arms towards thesite, secretes an adhesive, and attaches temporarily. Thenit alternatively attaches and detaches each brachiolar armand ‘‘walks’’ along the substratum. The brachiolar armsare morphologically similar to the tube feet of the adultsea star: they both consist of a proximal flexible stem anda distal sensory-secretory area (Haesaerts et al., 2005a, b).These morphological similarities (as well as the similaritiesin the biochemical content of the adhesive systems describedin Section II) between brachiolaria larvae and adult sea starspossibly reflect an economy at the genetic level (Haesaertset al., 2005b).

Turbulent channel flow measurements showed that shearstresses of about 1 Pa can dislodge temporarily attachedbrachiolar arms. This value is surprisingly low comparedto typical stresses in a wave-swept environment, which canreach up to 10 Pa. However, the value of the shear stress inthe channel flow does not correspond to field conditionsbecause the applied water channel creates a turbulentboundary layer with a steady overall flow rate, whereas in thefield high shear stresses last only few seconds and the forcesapplied on the larvae are reduced due to the roughness of thesubstratum and the neighbouring organisms (Haesaerts et al.,2005a). Organisms can be more easily dislodged by lastingsteady shear stresses than by the high but brief peaks, whichcan explain why brachiolar larvae seem to sustain muchhigher shear stresses in the field than in the water channel.

( iv ) Threads for restricted walking. During the pre-metamorphic stage, some larvae adopt a strategy of restrictedwalking: they stick to the substratum by means of flexiblethreads, so that they can move and explore whether thesurface is suitable for settlement, while remaining looselyattached to it. In contrast to the other surface structuresdescribed here, threads are extracellular. Cyphonautes,a larval form of the bryozoan Membranipora membranacea,for example, secrete mucus threads during exploration,

which attach to the substratum, while the larva keepsmoving around (Atkins, 1955); the exact adhesive mechanismremains unknown. It is interesting that cyphonautes explorehabitats in all directions and often upstream (Abelson, 1997).This is a rather unusual property considering that larvaemostly tend to settle in areas of low hydrodynamic stress,where they are less likely to be washed away (e.g. Abelson,1997; Abelson & Denny, 1997; Koehl & Hadfield, 2004; seealso Koehl, 2007, for a review).

A similar mechanism of attachment was studied byAbelson, Weihs & Loya (1994) in coral larvae, which producemucous threads up to 100 body lengths in size and are ableto settle in environments of high food-particle fluxes and lowsedimentation rates. Other larvae using threads include thedoliolaria larvae of crinoids, described in Section III.2a, andthe cyprid larvae of barnacles (Fig. 6C), described in SectionIII.2b.

(b) Branched-feet projecting morphologies

Branched-feet projecting morphologies consist of a mainflexible trunk with side and terminal branches; theapices of the terminal branches have one or moresensory/secretory/adhesive apices. Often located aroundthe mouth, most branched-feet projecting morphologiesare passively exposed to the water and used for funneland suspension feeding. Here we focus on branched-feetprojecting morphologies used in functions requiring strongeradhesion than funnel or suspension feeding. Three types ofsuch podia are relevant: digitate, peltate, and pinnate.

( i ) Digitate podia. Digitate podia are cylindrical witha slender tip (Fig. 7A). They occur in ophiuroids andcrinoids, and due to their histological similarities to tube feet,Flammang (1996) classified them as tube-foot morphotypes.In their studies on the crinoid Antedon bifida, Nichols (1960)and Lahaye & Jangoux (1985) also characterised digitatepodia as tube feet. The shape of digitate podia is similar

1 mm

A B C

Fig. 7. Branched-feet projecting morphologies. (A) Digitate podia of the ophiuroid Ophiothrix fragilis (Santos et al., 2009). (B) Peltatebuccal tentacle of the aspidochirotid holothurian Paroriza prouhoi. c: ciliates, d: digits (Roberts & Moore, 1997). (C) Pinnate tentaclesin the holothurian Synapta maculata P: pinnule, S: stem (Flammang & Conand, 2004). The insets correspond to the structural levelillustrations in Fig. 1.



to the knob-ending podia described in Section IV.2ai, butdigitate podia are smaller (see Figs 5B, 7A) and belong to abranched morphology. Digitate podia are mostly involvedin catching particles and filter feeding (Pentreath, 1970;Warner, 1982), functions which may explain why adhesivepapillae are discretely distributed, whereas in disc-endingand knob-ending tube feet the adhesive area is large andcontinuous, allowing strong attachment for locomotion orfor maintaining position (Santos et al., 2009).

( ii ) Peltate and pelto-digitate tentacles. Peltate tentaclesoccur in aspidochirotid holothurians and elasipods, whichare both deposit feeders. The basic shape of these tentaclesis cauliflower-like (Fig. 7B), but a large number of function-related variations exist (Hudson et al., 2003). The elasipodLaetmogone violacea, for example, has peltate podia with alarge adhesive surface area that enables sweeping off thesediment. The aspidochirotid Bathyplotes natans, on the otherhand, has finer peltate-like tentacles, with a higher degreeof branching that enables the selection of high-qualitydetrital particles without taking up much sediment. The largeaspidochirotid Paroriza pallens is equipped with tentacles thatare intermediate between peltate and digitate (pelto-digitate;Roberts & Moore, 1997). Pelto-digitate tentacles allow theanimal to plough the seabed ingesting large quantities ofsediment (Hudson et al., 2003). Other elasipodid as wellas aspidochirotid holothurians with peltate or pelto-digitatetentacles are described in Roberts & Moore (1997).

( iii ) Pinnate tentacles. Although apodid holothurians aredeposit feeders, they are not equipped with peltate tentacles,but with pinnate tentacles. Roberts (1982) argued thatpinnate tentacles are more efficient than peltate tentacles. Anapodous holothuroid with pinnate podia is Synapta maculata,a holothuroid that reaches lengths of 5 m – considerablylarger than other holothurians which have a typical length of20 cm, the smallest are less than 1 cm. The pinnate tentaclesof Synapta maculata are more prehensile than those of otherapodid holothurians, enabling them to wrap sea grass leavesor press against soft surfaces (Flammang & Conand, 2004)(Fig. 7C). The external epidermis of the tentacles containsa duo-gland system with adhesive secretions employed forcollecting and handling particles and de-adhesive secretionsfacilitating the release of the particles into the mouth.

(c) Tentacular projecting morphologies

For capturing a moving target (prey or a mate) orimmobilizing a predator, benthic animals use tentacles thatfirst mechanically immobilise the target and then release anadhesive to create an attachment. The mechanisms canbe divided into two groups: autotomising tentacles andretracting tentacles. The former are used to immobilise apredator and then autotomise, allowing the intended prey tocrawl away. The latter are employed for feeding and matingand are pulled back into the animal’s body after use.

( i ) Autotomising tentacles. The most well-known mecha-nism of autotomising tentacles is that of Cuvierian tubules(Figs. 2C, 8A). Apart from the unique character of thesecreted adhesive described in Section II.2b, the adhesive

strength of Cuvierian tubules depends on the morpholog-ical and mechanical properties of their inner tissues. Forexample, Flammang et al. (2002) found up to three timeshigher tenacities for shear loadings compared to peeling andsuggested that the difference in tenacity for both directionscan be related to function: ‘‘The higher tenacities associ-ated with shear loading and preliminary compression of thetubules indicate that Cuvierian tubules are well tailored forfunctioning as adhesive defense organs. Indeed, in nature, apotential predator entangled in Cuvierian tubules will mostlikely apply shear loads on the tubules and will thereforeexperience high adhesion strengths. Moreover, by pulling onthe tubules when trying to free itself, it will also maximizetheir stickiness.’’ (Flammang et al., 2002, p. 1114).

( ii ) Retracting tentacles. Tentacular projecting morpholo-gies can also be used for capturing prey or mates. Thetentacles are retracted into the animal’s body after use, buttheir structure and mechanisms strongly resemble that ofautotomising tentacles.

Retracting tentacles have been found in the cephalopodNautilus sp. Cephalopods usually use suckers for attachmentbut a few families of this class use adhesive secretionsinstead (see Von Byern & Klepal, 2006, for a review).Nautilus sp. has numerous tentacles without suckers or armhooks to attach to a substratum, capture prey, or clingto other individuals during mating. Instead, the tentaclessecrete adhesive granules that contain carbohydrates andproteins (Barber & Wright, 1969; Kier, 1987; Muntz &Wentworth, 1995).

Clione limacina (a pteropod mollusc) possesses a fast-strike feeding tentacular apparatus (studied by Hermans& Satterlie, 1992), which resembles the tentacles of Nautilus

sp. In the presence of a prey item, the mollusc opens itsmouth within 10–20 ms; then, in 50–70 ms, three pairsof oral appendages, called buccal cones, are hydrostaticallyinflated, extruded, surround the prey, and release a viscousmaterial that adheres to the shell of the prey (Fig. 8B).A third mechanism of retracting tentacles is that of Coeloplana

bannworthi (Ctenophora, Platyctenida), described in SectionsII.2b and III.2a.

V. DISCUSSION

We reviewed the temporary adhesion of benthic animalsfrom the perspective of three structural levels: thebiochemical content of the adhesive secretions, the micro-and mesoscopic surface geometry and material properties,and the macroscopic external morphology of the adhesiveorgans. We classified the temporary adhesive systems of arange of diverse animals into subcategories of these threestructural levels as described in Fig. 1, demonstrating that alimited set of concepts holds for a wide range of temporaryadhesive systems, perhaps indicating convergent evolution.

In all reviewed cases, the adhesive secretions providebinding to the substratum at a molecular scale, whereasthe surface and material characteristics and the external



A B

Fig. 8. Tentacular projecting morphologies. (A) The holothuroid Bohadschia argus with Cuvierian tubules in view (courtesyP. Flammang). (B) The pteropod mollusc Clione limacina with mouth open and five of the six buccal cones protruded. bc: buccalcones, m: mouth, t: tentacles (Hermans & Satterlie, 1992). The insets correspond to the structural level illustrations in Fig. 1.

morphology increase the contact area with the substratumfrom micro- to macroscales. From a cost-effectiveness point-of-view, a minimal amount of adhesive should be secreted,particularly for common activities such as locomotion andfeeding. This can be achieved by means of strongly adhesivemolecules, but also by the adaptability of a surface. It isalso possible that a trade-off exists between the softness of thematerial which the adhesive organ is made of and the amountof the secreted adhesive, because an excessively soft materialwould not be functional either. This trade-off may be furtherdependent on the action for which adhesion is used. Forinfrequent actions that are critical for the animal’s survival,such as predator immobilisation, surface optimisation mightbe less critical and larger amounts of adhesives must be used.This could be why we found more refined and elaboratesurface structures and adaptive morphologies for commonfunctions such as feeding activities than for occasional onessuch as predator defence.

(1) Biological research

Using our perspective of the three structural levels, key areasdeserving further investigation can be identified.

There is mixed information on the performance ofadhesive secretions on substrata of different polarity.Hennebert et al. (2008) reported that the footprints left by thesea star Asterias rubens on glass, mica, and Teflon had identicalmorphology, shape, and diameter, and that only the quantityof the secreted adhesive was larger for hydrophilic than forhydrophobic surfaces. Santos & Flammang (2006), on theother hand, found similar tenacities of Paracentrotus lividus

on glass, polypropylene, and polystyrene, but higher valueson poly(methyl metacrylate). Moreover, ‘‘abiotic’’ surfacesare usually covered with algal, microbial, or bacterial filmsand the effect of these organic films on adhesion is notknown. A number of studies suggested that organic filmsenhance larval settlement (e.g. Brancato & Woollacott, 1982;O’Connor & Richardson, 1998), whereas others questionedsuch facilitation (e.g. Keough & Raimondi, 1995; Robertset al., 1991). A more systematic investigation of adhesives

suitable for abiotic substrata may show whether these alsofulfil the requirements for adhesion to biotic substrata.

At the level of micro- and mesoscopic surface geometryand material properties, it is possible that cilia contributemechanically to adhesion by increasing the contact area.The contribution of cuticular villi to the adhesion ofcyprids by penetrating through the secreted adhesive toprovide direct contact with the substratum deserves furtherinvestigation, considering that this may represent an unusualcombination of wet and dry adhesion underwater. Moreover,it is possible that, due to their softness, papillae can adapt tothe irregularities of a surface, increasing the area of contactand therefore the adhesive capability of the animal.

Turning to the level of macroscopic external morphology,stem-disc structures are of particular interest in advancingour understanding of how the shape of an adhesiveorgan contributes to adhesion. Despite the apparentdifferences between various tube-foot morphotypes and theirpossible relations to function and habitat, no quantitativecomparisons have been made of the adhesive strength of thesedifferent morphotypes. According to Santos & Flammang(2005), the disc is important because it releases an adhesive,whereas connective tissue in the stem withstands the tensionsexerted by the hydrodynamic forces. Gorb & Varenberg(2007) suggested that a stem-disc morphology is a highlyfavourable shape for adhesion. They manufactured modelstem-disc geometries and found that they can sustain up toeight times larger pull-off forces than flat pillars can. Theyalso showed that there was a critical difference in the failuremode of flat pillars and of stem-disc geometries during thepulling: whereas in flat pillars contact separation started at anarbitrary point at the disc edge and propagated through thecentre to the opposite edge as a straight line, during pullingof stem-disc geometries a void was created in the centre ofthe disc and developed towards the disc edges. The strongeradhesion was due to the formation of this suction-like devicewhen pulling the flexible stem.

Quantitative data are lacking regarding low-curvaturemorphologies: one might expect that due to their relativelylarge size and ability to match the profile of a rough



surface, low-curvature morphologies should be smootherthan projecting morphologies. In Fig. 4A, it is clear, however,that the low-curvature ventral surface of Sepia tuberculata

consists of ridges at a microscale. A pattern of microscalegrooves can be also distinguished in Idiosepius sp. (Fig. 4B).This micropattern in Idiosepius sp. reminds the recentlydiscovered nanopattern on the apparent flat (i.e. low-curvature) tree frog toe pads (Scholz et al., 2009).

Future studies could investigate the movements accompa-nying the attachment and detachment of benthic animals.Even for tube feet, which are considered passive attach-ment devices, the activation of muscles and/or direction ofmacroscopic movements may be critical for adhesion andde-adhesion.

Finally, little is known about the cost of adhesion in benthicanimals. Donovan & Carefoot (1997) estimated the energycost of adhesive crawling for the abalone Haliotis kamtschatkana

(transitory adhesion) and VandenSpiegel et al. (2000)evaluated the efficiency of the regeneration mechanism ofCuvierian tubules in the sea cucumber Holothuria forskali. Byinvestigating the cost of temporary adhesion mechanisms inbenthic animals, we can gain a better understanding aboutthe trade-offs involved.

(2) Comparative studies

Multivariate statistics may be used to test our categorisationof adhesion mechanisms. Multivariate statistics have provenof great use in comparative biology (Felsenstein, 1985),particularly when morphological data are involved (Adams,Rohlf & Slice, 2004; Koehl, 1996). Zani (2000) statisticallyanalysed independent contrasts of claw and toe morphologyand of clinging performance from 85 lizard taxa and foundthat increased claw curvature, toe width, and numberof adhesive lamellae correlated with increased clingingperformance on smooth substrata, whereas increased clawheight and decreased toe length correlated with increasedclinging performance on rough substrata. Other studies haveinvestigated correlations between an animal’s morphologyand habitat (Irschick et al., 2005; Melville & Swain,2000), or changes in morphology due to evolutionarychanges of function (Schulte et al., 2004; Verwaijen & VanDamme, 2007). Statistical analyses should be combined withquantitative mechanistic analyses (Koehl, 1996) in order tounderstand how function depends on and affects adhesionmechanisms at the three structural levels.

In this review, a number of adhesive systems were discussedin more than one section, indicating that their adhesionfunctions simultaneously at more than one structural level.Therefore, we propose measuring and modelling adhesionforces across scales, from molecular to macroscopic. Ingeckos extensive studies revealed that multiscale hierarchical(surface) structures allow these animals to grip substrata ofvarious roughnesses (Gao et al., 2004). It is likely that theadhesion of benthic animals is also the result of multiscaleeffects: surface viscoelasticity and structures initiate the pointsof contact with the substratum at a micro- and mesoscale, theadhesive secretions fill in surface irregularities at a nanoscale,

and external morphology provides support and/or increasescontact with the substratum at a macroscale. A self-evident but fundamental difference between the adhesivemechanisms of terrestrial animals, such as geckos, andof benthic animals is that the latter must displace waterto initiate contact with a surface. It is unknown whetherwater displacement occurs due to the surface properties ofan adhesive organ (e.g. hydrophobicity/hydrophilicity), thenature of their secretions, or results directly from jettisonaction of the animal’s musculature.

Little is known about the operating time scales oftemporary adhesives compared to other modes of temporaryadhesion such as suction or dry adhesion. Some benthicanimals such as Sepia spp. (see Section IV.1a) use suctionas an initial phase of attachment and secrete temporaryadhesives afterwards to secure fixation to the substratum.Limpets also use suction for short-term attachment (up to afew hours) and switch to glue-like adhesion when they haveto remain attached for longer periods (Smith, 1991, 1992).The tentacles of octopods and decapods only bear suckersand do not produce adhesives; still the fast-swimming, open-water decapods of the suborder Oegopsida are able toproduce pressure differentials up to 800 kPa (Kier & Smith,2002). The tentacles of Nautilus spp., on the other hand, lacksuckers and operate by means of adhesive secretions only.A comparative study of sucker-based, secretion-based, anddry attachment mechanisms and their operating time scalesmay reveal the conditions in which each of these mechanismsis most effective.

(3) Biomimetic potential

The environment of benthic animals has similarities tothe internal environment of mammals. Inside the humanbody, tissues and organs are covered or surrounded byfluids that resemble sea water with respect to pH andionic composition (Flammang et al., 2005). Consequently,the temporary adhesion mechanisms of benthic animalshave become a source of inspiration for the developmentof artificial biomedical adhesives (Albala, 2003; Lee et al.,2007; Ninan et al., 2003; Strausberg & Link, 1990; Tatehataet al., 2001).

Studies of geometry and morphology in terrestrial animalshave inspired the development of artificial dry adhesivesystems (Arzt, 2006; Crosby, Hageman & Duncan, 2005;Geim et al., 2003; Gorb et al., 2007; Gravish et al., 2010; Sitti& Fearing, 2003; Yurdumakan et al., 2005), whereas artificialadhesive systems inspired by the adhesion mechanisms ofbenthic animals have mainly been restricted to analysingand mimicking the biochemical content of such systems(Burzio et al., 1997; Deming, 1999; Taylor & Weir, 2000;see also Flammang et al., 2005, for a review). Clearly,optimisation of the chemical content of an adhesive isimportant, but developers of artificial adhesive systems forbiomedical applications should also consider whether theycan improve adhesion in other ways. For example, whetherflexible and extensible bonds between two adhering surfacesincrease adhesion, whether an adhesive could be patterned,



or whether multiple isolated adhesive regions are better thanan extended adhesive surface.

To obtain insight into the combined effects of morphology,geometry, and secretions in biomedical adhesive systems,Dodou, Del Campo & Arzt (2007) carried out a series ofin vitro experiments. Previous research (Dodou, Breedveld& Wieringa, 2006) showed that adhesion to soft tissuesinside the human body can be considerably increased byusing mucoadhesives, which can be incorporated into thefeet of an in vivo robot that walks along the surface of softtissues or inside hollow organs. Dodou et al. (2007) showedthat a patterned adhesive generated significantly higher gripas compared to non-patterned adhesive films. Non-adhesivepatterns yielded much lower friction values, in fact lower thanthe friction generated by non-adhesive flat surfaces. Thesefindings corroborate that adhesion in wet environments canbe the combined result of mechanisms across scales, frommolecular to macroscopic.

VI. CONCLUSIONS

(1) Temporary adhesion in benthic animals can beconsidered as a multiscale architectural problem ofmolecular bonds in combination with an adaptivegeometric match between the adhesive organ and thesubstratum from micro- to macroscopic levels.

(2) Disparate taxa share common adhesion mechanisms,perhaps indicating convergent evolution.

(3) Optimisation of the chemical content of an adhesive isimportant, but developers of artificial adhesive systemsshould also consider that adhesion can be improvedin other ways, namely by optimising the contact areathrough adaptive geometries.

VII. ACKNOWLEDGEMENTS

The research of D. Dodou is supported by the DutchTechnology Foundation STW, Applied Science Divisionof NWO and the Technology Program of the Ministry ofEconomic Affairs. We thank two anonymous reviewers forproviding many constructive comments which helped toimprove this paper considerably. We also thank the AssistantEditor for valuable amendments in the manuscript.

VIII. REFERENCES

Abelson, A. (1997). Settlement in flow: Upstream exploration of substrata by weaklyswimming larvae. Ecology 78, 160–166.

Abelson, A. & Denny, M. (1997). Settlement of marine organisms in flow. Annual

Review of Ecology and Systematics 28, 317–339.Abelson, A., Weihs, D. & Loya, Y. (1994). Hydrodynamic impediments to

settlement of marine propagules, and adhesive-filament solutions. Limnology and

Oceanography 39, 164–169.Adams, D. C., Rohlf, F. J. & Slice, D. E. (2004). Geometric morphometrics: Ten

years of progress following the ‘revolution’. Italian Journal of Zoology 71, 5–16.

Adams, P. J. M. & Tyler, S. (1980). Hopping locomotion in a nematode: Functionalanatomy of the caudal gland apparatus of Theristus caudasaliens sp. n. Journal of

Morphology 164, 265–285.Albala, D. M. (2003). Fibrin sealants in clinical practice. Cardiovascular Surgery 11 (S1),

5–11.Aldred, N. & Clare, A. S. (2008). The adhesive strategies of cyprids and development

of barnacle-resistant marine coatings. Biofouling 24, 351–363.Ameye, L., Hermann, R., Dubois, P. & Flammang, P. (2000). Ultrastructure of the

echinoderm cuticle after fast-freezing/freeze substitution and conventional chemicalfixations. Microscopy Research and Technique 48, 385–393.

Arzt, E. (2006). Biological and artificial attachment devices: Lessons for materialsscientists from flies and geckos. Materials Science and Engineering C 26, 1245–1250.

Arzt, E., Gorb, S. & Spolenak, R. (2003). From micro to nano contacts inbiological attachment devices. Proceedings of the National Academy of Sciences USA

100, 10603–10606.Atkins, D. (1955). The cyphonautes larvae of the Plymouth area and the

metamorphosis of Membranipora Membranacea (L.). Journal of the Marine Biological

Association of the UK 34, 441–449.Ball, B. & Jangoux, M. (1990). Ultrastructure of the tube foot sensory-secretory

complex in Ophiocomina nigra (Echinodermata, Ophiuridea). Zoomorphology 109,201–209.

Barber, V. C. & Wright, D. E. (1969). The fine structure of the sense organs of thecephalopod mollusc Nautilus. Zeitschrift fur Zellforschung und Mikroskopische Anatomie 102,293–312.

Blake, D. B. (1989). Asteroidea: Functional morphology, classification and phylogeny.In Echinoderm Studies (eds. M. Jangoux and J. M. Lawrence), pp. 179–223.Balkema, Rotterdam.

Bouland, C., Massin, C. & Jangoux, M. (1982). The fine structure of the buccaltentacles of Holothuria forskali (Echinodermata, Holothuroidea). Zoomorphology 101,133–149.

Brancato, M. S. & Woollacott, R. M. (1982). Effect of microbial films onsettlement of bryozoan larvae (Bugula simplex, B. stolonifera and B. turrita). Marine

Biology 71, 51–56.Burzio, L. O., Burzio, V. A., Silva, T., Burzio, L. A. & Pardo, J. (1997).

Environmental bioadhesion: Themes and applications. Current Opinion in Biotechnology

8, 309–312.Butman, C. A., Grassle, J. P. & Webb, C. M. (1988). Substrate choices made by

marine larvae settling in still water and in a flume flow. Nature 333, 771–773.Callow, M. E. & Callow, J. A. (2001). Marine biofouling: A sticky problem. Biologist

49, 1–5.Costelloe, J. (1988). Reproductive cycle, development and recruitment of two

geographically separated populations of the dendrochirote holothurian Aslia lefevrei.Marine Biology 99, 535–545.

Crosby, A. J., Hageman, M. & Duncan, A. (2005). Controlling polymer adhesionwith ‘‘pancakes’’. Langmuir 21, 11738–11743.

Deming, T. J. (1999). Mussel byssus and biomolecular materials. Current Opinion in

Chemical Biology 3, 100–105.DeMoor, S., Waite, J. H., Jangoux, M. & Flammang, P. (2003). Characterization

of the adhesive from Cuvierian tubules of the sea cucumber Holothuria forskali

(Echinodermata, Holothuroidea). Marine Biotechnology 5, 45–57.Dodou, D., Breedveld, P. & Wieringa, P. A. (2006). The role of geometry in the

friction generated on the colonic surface by mucoadhesive films. Journal of Applied

Physics 100, 015904.1–015904.13.Dodou, D., Del Campo, A. & Arzt, E. (2007). Mucoadhesive micropatterns for

enhanced grip with the colon. In Proceedings of the 29th Conference of the IEEE EMBS,pp. 1457–1462. Lyon, France.

Donovan, D. A. & Carefoot, T. H. (1997). Locomotion in the abalone Haliotis

kamtschatkana: Pedal morphology and cost of transport. The Journal of Experimental

Biology 200, 1145–1153.Eeckhaut, I., Flammang, P., Lo Bue, C. & Jangoux, M. (1997). Functional

morphology of the tentacles and tentilla of Coeloplana bannworthi (Ctenophora,Platyctenida), an ectosymbiont of Diadema setosum (Echinodermata, Echinoida).Zoomorphology 117, 165–174.

Endean, R. (1957). The Cuvierian tubules of Holothuria leucospilota. Quarterly Journal of

Microscopical Science 98, 455–472.Engster, M. S. & Brown, S. C. (1972). Histology and ultrastructure of the tube foot

epithelium in the phanerozonian starfish, Astropecten. Tissue & Cell 4, 503–518.Fankboner, P. V. (1978). Suspension-feeding mechanisms of the armoured sea

cucumber Psolus chitinoides Clark. Journal of Experimental Marine Biology and Ecology

31, 11–25.Felsenstein, J. (1985). Phylogenies and the comparative method. The American

Naturalist 125, 1–15.Fishelson, L. (1981). Observations on the moving colonies of the genus Tethya

(Demospongia, Porifera). I. Behaviour and cytology. Zoomorphology 98, 89–99.Flammang, P. (1996). Adhesion in echinoderms. In Echinoderm Studies (eds.

M. Jangoux and J. M. Lawrence), 5, pp. 1–60. Balkema, Rotterdam.



Flammang, P. (2006). Adhesive secretions in echinoderms: An overview. In Biological

Adhesives (A. M. Smith and J. A. Callow), pp. 183–206. Springer-Verlag, BerlinHeidelberg.

Flammang, P. & Conand, C. (2004). Functional morphology of the tentacles inthe apodid holothuroid Synapta maculata. In Echinoderms: Munchen (Proceedings of the

11th International Echinoderm Conference, Munich, Germany) (eds. T. Heinzeller and J.H. Nebelsick), pp. 327–333. Taylor & Francis Group, London.

Flammang, P. & Jangoux, M. (1992). Functional morphology of the locomotorypodia of Holothuria forskali (Echinodermata, Holothuroidea). Zoomorphology 111,167–178.

Flammang, P. & Jangoux, M. (1993). Functional morphology of coronaland peristomeal podia in Sphaerechinus granularis (Echinodermata, Echinoidea).Zoomorphology 113, 47–60.

Flammang, P. & Jangoux, M. (2004). Bioadhesion models from marine invertebrates:

An integrated study–biomechanical, morphological, biochemical, molecular–of the processes

involved in the adhesion of Cuvierian tubules in sea cucumbers (Echinodermata, Holothuroidea).Belgium: University of Mons-Hainaut. Retrieved from http://www.dtic.mil/cgi-bin/GetTRDoc?AD=ADA427068&Location=U2&doc=GetTRDoc.pdf

Flammang, P., Demeulenaere, S. & Jangoux, M. (1994). The role of podialsecretions in adhesion in two species of sea stars (Echinodermata). The Biological

Bulletin 187, 35–47.Flammang, P., De Ridder, C. & Jangoux, M. (1990). Morphologie fonctionnelle

des podia penicilles chez l’ echinide fouisseur Echinocardium cordatum (Echinodermata).In Echinoderm Research (eds. C. de Ridder, Ph. Dubois, M-Ch. Lahaye andM. Jangoux), pp. 233–238. Balkema, Rotterdam.

Flammang, P., Michel, A., Van Cauwenberge, A., Alexandre, H. & Jangoux,M. (1998). A study of the temporary adhesion of the podia in the sea star Asterias

rubens (Echinodermata, Asteroidea) through their footprints. The Journal of Experimental

Biology 201, 2383–2395.Flammang, P., Ribesse, J. & Jangoux, M. (2002). Biomechanics of adhesion in

sea cucumber Cuvierian tubules (Echinodermata, Holothuroidea). Integrative and

Comparative Biology 42, 1107–1115.Flammang, P., Santos, R. & Haesaerts, D. (2005). Echinoderm adhesive

secretions: From experimental characterization to biotechnological applications.In Progress in Molecular and Subcellular Biology, Subseries Marine Molecular Biotechnology

(ed. V. Matranga), pp. 201–220. Springer-Verlag, Berlin Heidelberg.Franc, J-M. (1978). Organization and function of ctenophore colloblasts:

An ultrastructural study. The Biological Bulletin 155, 527–541.Gao, H., Wang, X., Yao, H., Gorb, S. & Arzt, E. (2004). Mechanics of hierarchical

adhesion structures of geckos. Mechanics of Materials 37, 275–285.Geim, A. K., Dubonos, S. V., Grigorieva, I. V., Novoselov, K. S., Zhukov,

A. A. & Shapoval, S. Yu. (2003). Microfabricated adhesive mimicking geckofoot-hair. Nature Materials 2, 461–463.

Gorb, S. N. (2008). Biological attachment devices: Exploring nature’s diversity forbiomimetics. Philosophical Transactions of the Royal Society A 366, 1557–1574.

Gorb, S., Jiao, Y. & Scherge, M. (2000). Ultrastructural architecture andmechanical properties of attachment pads in Tettigonia viridissima (OrthopteraTettigoniidae). Journal of Comparative Physiology A 186, 821–831.

Gorb, S. N., Sinha, M., Peressadko, A., Daltorio, K. A. & Quinn, R. D. (2007).Insects did it first: A micropatterned adhesive tape for robotic applications.Bioinspiration & Biomimetics 2, S117–S125.

Gorb, S. N. & Varenberg, M. (2007). Mushroom-shaped geometry of contactelements in biological adhesive systems. Journal of Adhesion Science and Technology

21, 1175–1183.Gravish, N., Wilkinson, M., Sponberg, S., Parness, A., Esparza, N., Soto, D.,

Yamaguchi, T., Broide, M., Cutkosky, M., Creton, C. & Autumn, K. (2010).Rate-dependent frictional adhesion in natural and synthetic gecko setae. Journal of

the Royal Society Interface 7, 259–269.Haesaerts, D., Finlay, J. A., Callow, M. E., Callow, J. A., Grosjean, P.,

Jangoux, M. & Flammang, P. (2005a). Evaluation of the attachment strengthof individuals of Asterina gibbosa (Asteroidea, Echinodermata) during theperimetamorphic period. Biofouling 21, 229–235.

Haesaerts, D., Jangoux, M. & Flammang. P. (2005b). The attachment complex ofbrachiolaria larvae of the sea star Asterias rubens (Echinodermata): An ultrastructuraland immunocytochemical study. Zoomorphology 124, 67–78.

Hamwood, T. E., Cribb, B. W., Halliday, J. A., Kearn, G. C. & Whitting-ton, I. D. (2002). Preliminary characterisation and extraction of anterior adhesivesecretion in monogenean (platyhelminth) parasites. Folia Parasitologica 49, 39–49.

Hennebert, E., Viville, P., Lazzaroni, R. & Flammang, P. (2008). Micro- andnanostructure of the adhesive material secreted by the tube feet of the sea star Asterias

rubens. Journal of Structural Biology 164, 108–118.Hermans, C. O. (1983). The duo-gland adhesive system. Oceanography and Marine

Biology: An Annual Review 21, 283–339.Hermans, C. O. & Satterlie, R. A. (1992). Fast-strike feeding behavior in a

pteropod mollusk, Clione limacina Phipps. The Biological Bulletin 182, 1–7.Hudson, I. R., Wigham, B. D., Billett, D. S. M. & Tyler, P. A. (2003). Seasonality

and selectivity in the feeding ecology and reproductive biology of deep-sea bathyalholothurians. Progress in Oceanography 59, 381–407.

Irschick, D. J., Vanhooydonck, B., Herrel, A. & Meyers, J. (2005). Intraspecificcorrelations among morphology, performance and habitat use within a green anolelizard (Anolis carolinensis) population. Biological Journal of the Linnean Society 85, 211–221.

Jangoux, M. & Lahaye, M-Ch. (1990). The attachment complex of the doliolarialarvae of Antedon bifida (Echinodermata, Crinoidea). In Echninoderm Research (eds. C. deRidder, Ph. Dubois, M-Ch. Lahaye and M. Jangoux), pp. 99–105. Balkema,Rotterdam.

Kearn, G. C. & Evans-Gowing, R. (1998). Attachment and detachment of theanterior adhesive pads of the monogenean (platyhelminth) parasite Entobdella soleae

from the skin of the common sole (Solea solea). International Journal for Parasitology 28,1583–1593.

Keogh, J. K. & Keegan, B. F. (2006). Ultrastructure of the podia of Amphiura chiajei

and Amphiura filiformis and their role in feeding. Journal of the Marine Biological Association

of the UK 86, 817–822.Keough, M. J. & Raimondi, P. T. (1995). Responses of settling invertebrate larvae

to bioorganic films: Effects of different types of films. Journal of Experimental Marine

Biology and Ecology 185, 235–253.Kier, W. M. (1987). The functional morphology of the tentacle musculature of Nautilus

pompilius. In Nautilus (eds. W. B. Saunders and N. H. Landman), pp. 257–269.Plenum Publishing Corporation, New York.

Kier, W. M. & Smith, A. M. (2002). The structure and adhesive mechanism ofoctopus suckers. Integrative and Comparative Biology 42, 1146–1153.

Koehl, M. A. R. (1996). When does morphology matter? Annual Review of Ecology and

Systematics 27, 501–542.Koehl, M. A. R. (2007). Mini review: Hydrodynamics of larval settlement into fouling

communities. Biofouling 23, 357–368.Koehl, M. A. R. & Hadfield, M. G. (2004). Soluble settlement cue in slowly moving

water within coral reefs induces larval adhesion to surfaces. Journal of Marine Systems

49, 75–88.Lahaye, M. C. & Jangoux, M. (1985). Functional morphology of the podia and

ambulacral grooves of the comatulid crinoid Antedon bifida (Echinodermata). Marine

Biology 86, 307–318.Leddy, H. A. & Johnson, A. S. (2000). Walking versus breathing: Mechanical

differentiation of sea urchin podia corresponds to functional specialization. The

Biological Bulletin 198, 88–93.Lee, H., Dellatore, S. M., Miller, W. M. & Messersmith, P. B. (2007). Mussel-

inspired surface chemistry for multifunctional coatings. Science 318, 426–430.Martin, G. G. (1978). The duo-gland adhesive system of the archiannelids Protodrilus

and Saccocirrus and the turbellarian Monocelis. Zoomorphologie 91, 63–75.McKenzie, J. D. (1987). The ultrastructure of the tentacles of eleven species of

dendrochirote holothurians studied with special reference to the surface coats andpapillae. Cell and Tissue Research 248, 187–199.

McKenzie, J. D. (1988). Ultrastructure of the tentacles of the apodous holothurianLeptosynapta spp (Holothurioidea: Echninodermata) with special reference to theepidermis and surface coats. Cell and Tissue Research 251, 387–397.

Melville, J. & Swain, R. (2000). Evolutionary relationships between morphology,performance and habitat openness in the lizard genus Niveoscincus (Scincidae:Lygosominae). Biological Journal of the Linnean Society 70, 667–683.

Mooi, R. (1986a). Non-respiratory podia of clypeasteroids (Echinodermata,Echinoides): I. Functional anatomy. Zoomorphology 106, 21–30.

Mooi, R. (1986b). Structure and function of clypeasteroid miliary spines(Echinodermata, Echinoides). Zoomorphology 106, 212–223.

Moynihan, M. (1983a). Notes on the behavior of Euprymna scolopes (Cephalopoda:Sepiolidae). Behaviour 85, 25–41.

Moynihan, M. (1983b). Notes on the behavior of Idiosepius pygmaeus (Cephalopoda;Idiosepiidae). Behaviour 85, 42–57.

Mullineaux, L. S. & Butman, C. A. (1991). Initial contact, exploration andattachment of barnacle (Balanus amphitrite) cyprids settling in flow. Marine Biology

110, 93–103.Muntz, W. R. A. & Wentworth, S. L. (1995). Structure of the adhesive surface of

the digital tentacles of Nautilus pompilius. Journal of the Marine Biological Association of the

UK 75, 747–750.Nakano, H., Hibino, T., Oji, T., Hara, Y. & Amemiya, S. (2003). Larval stages of

a living sea lily (stalked crinoid echinoderm). Nature 421, 158–160.Nichols, D. (1959). The histology and activities of the tube-feet of Echinocyamus

pusillus. Quarterly Journal of Microscopical Science 100, 539–555.Nichols, D. (1960). The histology and activities of the tube-feet of Antedon bifida.

Quarterly Journal of Microscopical Science 101, 105–117.Ninan, L., Monahan, J., Stroshine, R. L., Wilker, J. J. & Shi, R. (2003).

Adhesive strength of marine mussel extracts on porcine skin. Biomaterials 24,4091–4099.

O’Connor, N. J. & Richardson, D. L. (1998). Attachment of barnacle (Balanus

amphitrite Darwin) larvae: Responses to bacterial films and extracellular materials.Journal of Experimental Marine Biology and Ecology 226, 115–129.

Pawlik, J. R., Butman, C. A. & Starczak, V. R. (1991). Hydrodynamic facilitationof gregarious settlement of a reef-building tube worm. Science 251, 421–424.



Peattie, A. M. & Full, R. J. (2007). Phylogenetic analysis of the scaling of wet anddry biological fibrillar adhesives. Proceedings of the National Academy of Sciences USA 104,18595–18600.

Pentreath, R. J. (1970). Feeding mechanisms and the functional morphology ofpodia and spines in some New Zealand ophiuroids (Echinodermata). Journal of

Zoology 161, 395–429.Peressadko, A. & Gorb, S. N. (2004). When less is more: Experimental evidence

for tenacity enhancement by division of contact area. The Journal of Adhesion 80,247–261.

Phang, I. Y., Aldred, N., Clare, A. S. & Vancso, G. J. (2008). Towards ananomechanical basis for temporary adhesion in barnacle cyprids (Semibalanus

balanoides). Journal of the Royal Society Interface 5, 397–401.Phang, I. Y., Aldred, N., Ling, X. Y., Huskens, J., Clare, A. S. & Vancso, G. J.

(2010). Atomic force microscopy of the morphology and mechanical behaviour ofbarnacle cyprid footprint proteins at the nanoscale. Journal of the Royal Society Interface

7, 285–296.Roberts, D. (1982). Classification and the holothurian tentacle. In Echinoderms

(ed. J. M. Lawrence), pp. 117–121. Balkema, Rotterdam.Roberts, D. & Moore, H. M. (1997). Tentacular diversity in deep-sea deposit-

feeding holothurians: Implications for biodiversity in the deep sea. Biodiversity and

Conservation 6, 1487–1505.Roberts, D., Rittschof, D., Holm, E. & Schmidt, A. R. (1991). Factors

influencing initial larval settlement: Temporal, spatial and surface molecularcomponents. Journal of Experimental Marine Biology and Ecology 150, 203–221.

Roeleveld, M. A. C. (1972). A review of the Sepiidae (Cephalopoda) of SouthernAfrica. Annals of the South African Museum 59, 193–313.

Santos, R. & Flammang, P. (2005). Morphometry and mechanical design of tubefoot stems in sea urchins: A comparative study. Journal of Experimental Marine Biology

and Ecology 315, 211–223.Santos, R. & Flammang, P. (2006). Morphology and tenacity of the tube foot disc

of three common European sea urchin species: A comparative study. Biofouling 22,173–186.

Santos, R., Gorb, S., Jamar, V. & Flammang, P. (2005a). Adhesion of echinodermtube feet to rough surfaces. The Journal of Experimental Biology 208, 2555–2567.

Santos, R., Haesaerts, D., Jangoux, M. & Flammang, P. (2005b). Comparativehistological and immunohistochemical study of sea star tube feet (Echinodermata,Asteroidea). Journal of Morphology 263, 259–269.

Santos, R., Hennebert, E., Coelho, A. V. & Flammang, P. (2009). Theechinoderm tube foot and its role in temporary underwater adhesion. In Functional

Surfaces in Biology, Vol. 2 (ed. S. N. Gorb), pp. 9–41. Springer Science+BusinessMedia B.V.

Scherge, M. & Gorb, S. N. (2001). Biological Micro- and Nanotribology: Nature’s Solutions.Springer, Berlin.

Scholz, I., Barnes, W. J. P., Smith, J. M. & Baumgartner, W. (2009).Ultrastructure and physical properties of an adhesive surface, the toe pad epitheliumof the tree frog, Litoria caerulea White. The Journal of Experimental Biology 212, 155–162.

Schulte, J. A., II, Losos, J. B., Cruz, F. B. & Nunez, H. (2004). The relationshipbetween morphology, escape behaviour and microhabitat occupation in the lizardclade Liolaemus (Iguanidae: Tropidurinae∗: Liolaemini). Journal of Evolutionary Biology

17, 408–420.Shears, J. (1988). The use of a sand-coat in relation to feeding and diel activity in the

sepiolid squid Euprymna scolopes. Malacologia 29, 121–133.Singley, C. T. (1982). Histochemistry and fine structure of the ectodermal epithelium

of the sepiolid squid Euprymna scolopes. Malacologia 23, 177–192.Sitti, M. & Fearing, R. S. (2003). Synthetic gecko foot-hair micro/nano-structures

as dry adhesives. Journal of Adhesion Science and Technology 17, 1055–1073.Smith, A. B. (1978). A functional classification of the coronal pores of regular

echinoids. Palaeontology 21, 759–789.Smith, A. M. (1991). The role of suction in the adhesion of limpets. The Journal of

Experimental Biology 161, 151–169.

Smith, A. M. (1992). Alternation between attachment mechanisms by limpets in thefield. Journal of Experimental Marine Biology and Ecology 160, 205–220.

Smith, A. M. & Morin, M. C. (2002). Biochemical differences between trail mucusand adhesive mucus from marsh periwinkle snails. The Biological Bulletin 203,338–346.

Strausberg, R. L. & Link, R. P. (1990). Protein-based medical adhesives. Trends in

Biotechnology 8, 53–57.Tatehata, H., Mochizuki, A., Ohkawa, K., Yamada, M. & Yamamoto, H.

(2001). Tissue adhesive using synthetic model adhesive proteins inspired by themarine mussel. Journal of Adhesion Science and Technology 15, 1003–1013.

Taylor, C. M. & Weir, C. A. (2000). Synthesis of the repeating decapeptide unit ofMefp1 in orthogonally protected form. Journal of Organic Chemistry 65, 1414–1421.

Telford, M. & Mooi, R. (1986). Resource partitioning by sand dollars in carbonateand siliceous sediments: Evidence from podial and particle dimensions. The Biological

Bulletin 171, 197–207.Telford, M., Mooi, R. & Harold, A. S. (1987). Feeding activities of two species of

Clypeaster (Echinoides, Clypeasteroida): Further evidence of clypeasteroid resourcepartitioning. The Biological Bulletin 172, 324–336.

Thomas, L. A. & Hermans, C. O. (1985). Adhesive interactions between the tubefeet of a starfish, Leptasterias hexactis, and substrata. The Biological Bulletin 169,675–688.

Tyler, S. (1988). The role of function in determination of homology andconvergence – Examples from invertebrate adhesive organs. Fortschritte der Zoologie

36, 331–347.Tyler, S. & Rieger, G. E. (1980). Adhesive organs of the gastrotricha. I. Duo-gland

organs. Zoomorphologie 95, 1–15.VandenSpiegel, D. & Jangoux, M. (1987). Cuvierian tubules of the holothuroid

Holothuria forskali (Echinodermata): A morphofunctional study. Marine Biology 96,263–275.

VandenSpiegel, D., Jangoux, M. & Flammang, P. (2000). Maintaining the line ofdefense: Regeneration of Cuverian tubules in the sea cucumber Holothuria forskali

(Echinodermata, Holothuroidea). The Biological Bulletin 198, 34–49.Verwaijen, D. & Van Damme, R. (2007). Does foraging mode mould morphology

in lacertid lizards? Journal of Evolutionary Biology 20, 1950–1961.Vickery, M. S. & McClintock, J. B. (2000). Comparative morphology of tube feet

among the Asteroidea: Phylogenetic implications. American Zoology 40, 355–364.Von Byern, J. & Klepal, W. (2006). Adhesive mechanisms in cephalopods: A review.

Biofouling 22, 329–338.Von Byern, J., Rudoll, L., Cyran, N. & Klepal, W. (2008). Histochemical

characterization of the adhesive organ of three Idiosepius spp. species. Biotechnic

& Histochemistry 83, 29–46.Walker, G. & Yule, A. B. (1984). Temporary adhesion of the barnacle cyprid: The

existence of an antennular adhesive secretion. Journal of the Marine Biological Association

of the United Kingdom 64, 679–686.Warner, G. (1982). Food and feeding mechanisms: Ophiuroidea. In Echinoderm

Nutrition (eds. M. Jangoux and J. M. Lawrence), pp. 161–184. Balkema,Rotterdam.

Whittington, I. D., Armstrong, W. D. & Cribb, B. W. (2004). Mechanismof adhesion and detachment at the anterior end of Neoheterocotyle rhinobatidis

and Troglocephalus rhinobatidis (Monogenea: Monopisthocotylea: Monocotylidae).Parasitology Research 94, 91–95.

Whittington, I. D. & Cribb, B. W. (2001). Adhesive secretions in thePlatyhelminthes. Advances in Parasitology 48, 101–224.

Wiegemann, M. & Watermann, B. (2003). Peculiarities of barnacle adhesive curedon non-stick surfaces. Journal of Adhesion Science and Technology 17, 1957–1977.

Yurdumakan, B., Raravikar, N. R., Ajayan, P. M. & Dhinojwala, A.(2005). Synthetic gecko foot-hairs from multiwalled carbon nanotubes. Chemical

Communications, 3799–3801.Zani, P. A. (2000). The comparative evolution of lizard claw and toe morphology and

clinging performance. Journal of Evolutionary Biology 13, 316–325.


Mechanisms of temporary adhesion in benthic animals

Documents