The Musculotendinous System of an Anguilliform Swimmer ...glauder//reprints_unzipped/Danos et...

The Musculotendinous System of an AnguilliformSwimmer: Muscles, Myosepta, Dermis, and TheirInterconnections in Anguilla rostrata

Nicole Danos,1* Nina Fisch,2 and Sven Gemballa2

1Department of Organismic and Evolutionary Biology, Harvard University, Cambridge, Massachusetts 021382Department of Zoology, University of Tubingen, Auf der Morgenstelle 28, Tubingen D-72076, Germany

ABSTRACT Eel locomotion is considered typical of theanguilliform swimming mode of elongate fishes and hasreceived substantial attention from various perspectivessuch as swimming kinematics, hydrodynamics, musclephysiology, and computational modeling. In contrast tothe extensive knowledge of swimming mechanics, thereis limited knowledge of the internal body morphology,including the body components that contribute to thisfunction. In this study, we conduct a morphological anal-ysis of the collagenous connective tissue system, i.e., themyosepta and skin, and of the red muscle fibers thatsustain steady swimming, focusing on the interconnec-tions between these systems, such as the muscle-tendonand myosepta-skin connections. Our aim is twofold: (1)to identify the morphological features that distinguishthis anguilliform swimmer from subcarangiform and car-angiform swimmers, and (2) to reveal possible pathwaysof muscular force transmission by the connective tissuein eels. To detect gradual morphological changes alongthe trunk we investigated anterior (0.4L), midbody(0.6L), and posterior body positions (0.75L) using micro-dissections, histology, and three-dimensional reconstruc-tions. We find that eel myosepta have a mediolaterallyoriented tendon in each the epaxial and hypaxial regions(epineural or epipleural tendon) and two longitudinallyoriented tendons (myorhabdoid and lateral). The lattertwo are relatively short (4.5–5% of body length) andremain uniform along a rostrocaudal gradient. The skinand its connections were additionally analyzed usingscanning electron microscopy (SEM). The stratum com-pactum of the dermis consists of �30 layers of highly or-dered collagen fibers of alternating caudodorsal andcaudoventral direction, with fiber angles of 60.51 67.058 (n 5 30) and 57.58 6 6.928 (n 5 30), respectively.Myosepta insert into the collagenous dermis via fiberbundles that pass through the loose connective tissue ofthe stratum spongiosum of the dermis and either weaveinto the layers of the stratum compactum (weaving fiberbundles) or traverse the stratum compactum (transversefiber bundles). These fiber bundles are evenly distrib-uted along the insertion line of the myoseptum. Redmuscles insert into lateral and myorhabdoid myoseptaltendons but not into the horizontal septum or dermis.Thus, red muscle forces might be distributed along thesetendons but will only be delivered indirectly into thedermis and horizontal septum. The myosepta-dermisconnections, however, appear to be too slack for efficientforce transmission and collagenous connections betweenthe myosepta and the horizontal septum are at obtuseangles, a morphology that appears inadequate for

efficient force transmission. Though the main modes ofundulatory locomotion (anguilliform, subcarangiform,and carangiform) have recently been shown to be verysimilar with respect to their midline kinematics, we areable to distinguish two morphological classes withrespect to the shape and tendon architecture of myo-septa. Eels are similar to subcarangiform swimmers(e.g., trout) but are substantially different from carangi-form swimmers (e.g., mackerel). This information, inaddition to data from kinematic and hydrodynamic stud-ies of swimming, shows that features other than midlinekinematics (e.g., wake patterns, muscle activation pat-terns, and morphology) might be better for describingthe different swimming modes of fishes. J. Morphol.269:29–44, 2008. � 2007 Wiley-Liss, Inc.

KEY WORDS: myoseptal tendons; red muscle; skin;connective tissue; Anguilla rostrata

All swimming vertebrates share a similar axialmusculoskeletal morphology, consisting of a com-pression–resistant vertebral column or notochordsurrounded by axial musculature. The skeletaland muscular components interact through a com-plex three-dimensional system of connective tissue.In fishes the muscular system is organized as aseries of three-dimensionally folded segments, themyomeres. Adjacent myomeres are separated bysheets of connective tissue, the myosepta, intowhich the muscle fibers insert (Alexander, 1969;Gemballa and Vogel, 2002). Medially, the series ofmyomeres and myosepta inserts on the collagenousvertical septum and the bony axial skeleton. Later-ally, the whole system is wrapped by the skin. Inall basal groups of notochordates, including all fish

Contract grant sponsors: Society for Experimental Biology, Journalof Experimental Biology, Sigma-Xi, Wilhelm-Schuler-Stiftung andStrukturfond Baden-Wurttemberg at the University of Tubingen.

*Correspondence to: Nicole Danos, Department of Organismicand Evolutionary Biology, Harvard University, 26 Oxford Street,Cambridge, MA 02138. E-mail: [email protected]

Published online 21 September 2007 inWiley InterScience (www.interscience.wiley.com)DOI: 10.1002/jmor.10570

JOURNAL OF MORPHOLOGY 269:29–44 (2008)

� 2007 WILEY-LISS, INC.

taxa, the skin includes a multilayer system of col-lagenous connective tissue, the stratum compac-tum (Gemballa and Bartsch, 2002).

A coordinated and sequential activation of themyomeric musculature in most fishes generatesaxial undulatory swimming characterized bywaves of propulsive body undulations that traveldown the body towards the caudal fin while gradu-ally increasing in amplitude. However, we stillhave little understanding of how muscular activityis translated into undulatory swimming waves inmost fishes. This lack of understanding is mainlydue to the complexity of the components thatform the swimming apparatus. Muscle fibers areoriented in complex three-dimensional patternswithin myomeres (Alexander, 1969; Gemballa andVogel, 2002) and myosepta consist of a three-dimensional network of tendinous structures(Gemballa et al., 2003a). In addition, the bodyenvelope with its highly ordered layers of collagenfibers in the stratum compactum has been positedto affect whole body mechanics by acting as a ten-don transmitting forces produced by the axialmusculature to the tail, by stiffening the bodywhen under tension, or by storing elastic energyduring axial undulations (Wainwright et al., 1978;Hebrank, 1980; Hebrank and Hebrank, 1986;Alexander, 1987; Long et al., 1996; Brainerd andPatek, 1998). Similar hypotheses have been pro-posed for the other connective tissues, such as thehorizontal septum and myoseptal tendons (e.g.,Westneat and Wainwright, 2001; Gemballa et al.,2006).

Because of the complexity of the individual com-ponents, attempts to understand the function ofthe whole system have mostly dealt with just onecomponent at a time, focusing on the myosepta,vertebral column, segmented musculature, orskin. However, any functional analysis or modelshould consider information on the morphologyof as many components of the swimming appara-tus as possible, including the interconnections bet-ween each component (e.g., muscle-tendon associa-tions or connections between myosepta and skin).For example, it is critical to know the associationsof muscle fibers and myoseptal tendons if one is toanalyze the transmission of muscular forces. Arecent study that has linked muscle and tendonmorphology to swimming biomechanics by concur-rently analyzing swimming kinematics, muscle dy-namics, and myoseptal anatomy has led to animproved understanding of how muscle activity isconvergently translated into the uniform swim-ming in lamnid sharks and tunas (Donley et al.,2004, 2005; Gemballa et al., 2006). Other studieshave provided further morphological informationof the geometry of myoseptal tendons or of mus-cle-tendon associations (Gemballa and Vogel,2002; Gemballa and Treiber, 2003; review: Shad-wick and Gemballa, 2006). Such information has

provided input to dynamic models that describethe swimming movements of subcarangiform orcarangiform swimmers (Long et al., 2002). Theseexamples show that the study of fish swimmingcan benefit significantly from further morphologi-cal information.

One of the most characteristic ways of perform-ing undulatory swimming, anguilliform swimming,is displayed by many elongate or eel-like fishes.The anguilliform mode was described as a distinctswimming mode characterized by body waves withrelatively short wavelengths and large amplitudes(Lindsey, 1978). Most of the work on anguilliformswimming was carried out on the genus Anguillaand has not only covered kinematics but also thehydrodynamics, muscle dynamics, whole bodymechanics, and modeling (Gray, 1933a,b,c, 1968;Gillis, 1996, 1998a,b; Long, 1998; D’Aout et al.,2001; Muller et al., 2001; Tytell, 2004a,b; Tytelland Lauder, 2004; Kern and Koumoutsakos, 2006).However, recent work suggests that the dorsalmidline swimming profiles, from which kinematicvariables such as body wave wavelength and am-plitude are calculated, do not differ substantiallybetween anguilliform and subcarangiform or evencarangiform swimmers (Donley and Dickson, 2000;Lauder and Tytell, 2006). In contrast to the largebody of knowledge in these experimental fields, weknow surprisingly little about the underlying mor-phology of elongate fishes. If the kinematic profilesof fishes so different in external morphology are sosimilar, how does their internal locomotory anat-omy compare?

In this study, we address several aspects of themorphology of the swimming apparatus of theAmerican Eel, Anguilla rostrata. Our morphologi-cal analysis includes the red muscle fibers, thepart of the muscular system that powers steadyswimming, and the collagenous connective tissuesystem, i.e., myosepta and skin, that transmitsmuscular forces to the backbone. In particular wepay attention to the interconnections betweenthese systems, such as the muscle-tendon andmyosepta-skin connections. Our goal is twofold.First, we aim to identify the morphological fea-tures that distinguish this anguilliform swimmerfrom subcarangiform and carangiform swimmers.Such findings would facilitate the generation oftestable hypotheses on the locomotory function ofparticular structures. Second, we aim to revealpossible pathways of muscular force transmissionby the connective tissue in eels. Both skin andmyosepta have been implicated in force transmis-sion in some fishes including eels (Wainwrightet al., 1978; Hebrank, 1980; Wainwright, 1983;Hebrank and Hebrank, 1986; Westneat and Wain-wright, 2001). By describing the insertions of redmuscle fibers to collagen fiber tracts within myo-septa and skin, and the interconnections betweenmyoseptal tendons and the dermis, we aim to

30 N. DANOS ET AL.

Journal of Morphology DOI 10.1002/jmor

reevaluate the probability of these functionalhypotheses.

MATERIALS AND METHODSMaterial Examined

The results of this study are based on a collection of techni-ques applied to eight specimens of Anguilla rostrata Le Sueur,1817. Table 1 gives a list of specimens and summarizes theinvestigations applied to each.

Investigation of Myoseptal Shape andArchitecture

The techniques for studying myosepta and their architecturefollow the procedures described elsewhere (e.g., Gemballa andHagen, 2004) and are only briefly described here. The only tech-nical addition made here is the use of a computer-based 3D-reconstruction as an alternative source of three-dimensionaldata for myoseptal architecture and length measurements ofmyoseptal tendons.Data on the 3D-shape of myosepta were obtained from

cleared and double stained (for bone and cartilage according toDingerkus and Uhler, 1977) specimens of Anguilla rostrata thathad been skinned prior to clearing (Table 1; Specimens 1, 2). Af-ter completion of the staining procedure, the specimens weretransferred stepwise into pure ethanol in order to better visual-ize the connective tissue. For the following microdissections weused fine iris spring scissors (FST Vannas Mini) and removedeach myoseptum close to its insertion line along the vertebralaxis and along the horizontal and vertical septa, to examine thecollagen tracks and tendons within each myoseptum. Werecorded the anteriormost position of the insertion line on thevertebral column of each myoseptum and defined this positionas the axial position of the myoseptum (axial position 0.0L istip of snout; 1.0L is tip of caudal fin). Excised myosepta werekept separately in ethanol for further investigation of collagen

fiber tracts. A few myosepta were retained in their native axialposition. Their 3D-shape including attachment to the vertebralaxis was drawn using a camera lucida.

In addition to microdissections, we used digital 3D recon-structions (Amira v. 3.1) of three myosepta from horizontal orsagittal histological sections (anterior body position at 0.45L;midbody position at 0.60L; posterior body position at 0.75L; Ta-ble 1; Specimen 3). After manual alignment of a series of sec-tions with the alignment tools of this software we built a 3D-model of a complete myoseptum. We are confident that digitalreconstructions from histological sections were as reliable asour traditional microdissection technique outlined above sinceboth provided comparable results from different specimens.

In addition to the shape of myosepta we recorded the overallmyoseptal length. This was defined as the distance between thetip of the anteriorly-pointing and the posteriorly-pointing myo-septal cones (Shadwick and Gemballa, 2006). In the cleared andstained specimens (Table 1; Specimens 1, 2) this length wasmeasured to the closest tenth of a millimeter using a calipergauge. In the computer generated 3D-model (Table 1; Specimen3), we measured the overall myoseptal length by relating thescale bar of the original microscopic image to the Amira meas-uring tool.

Collagen fiber tracts of excised myosepta from Specimens 1and 2 were visualized and photographed by spreading them outon glass plates under polarized light (Zeiss Polarizer S and ana-lyzer A53 adapted to stereomicroscope; see Gemballa andHagen, 2004). In addition, collagen fiber tracts were recordedunder incident light directly from the myosepta that wereretained in the cleared and stained specimens. These tractswere included in the camera lucida drawings of myosepta (seeabove).

To visualize the spatial orientation of postcranial bony ele-ments, particularly the ossified myoseptal tendons (i.e., inter-muscular bones) and their relation to vertebrae, we ran highresolution l-3D X-ray CT scans using a RayScan 200 system(Walischmiller GmbH Germany; ARGE Metallguss at SteinbeisTransferzentrum at Fachhochschule Aalen, Germany). Scans ofa midbody segment of one specimen (Table 1; Specimen 4) wereobtained at a resolution of 25 lm (417 optical sections per verte-

TABLE 1. List of Anguilla rostrata specimens used in this study

Specimen no. [TL] PreparationBody regionexamined Investigation

1 [290 mm] Skinned, cleared and double stainedfor cartilage (Alcian Blue 8GX)and bone (Alizarin Red)

0.2–0.9L 3D-shape of myosepta, collagenous architectureof myosepta, measurement of overallmyoseptal length

2 [274 mm]

3 [295 mm] Horizontal sectionsa of: Computer based 3D reconstruction of myoseptafrom histological sections, measurement ofoverall myoseptal length and muscle fiberangles, insertion of RM into myoseptaltendons

Anterior body 0.42–0.48LMidbody (right side) 0.56–0.63LPosterior body 0.71–0.78LSagittal sectionsa (left side) 0.56–0.63L

3 [295 mm] Paraserial sectionsa 0.52–0.54L Interconnections between skin and myosepta,insertion of RM into myoseptal tendonsInterserial sectionsa 0.52–0.54L

4 [300 mm] Formalin fixed 0.55–0.58L 3D l-CT scan for bony elements5 [250 mm] Unskinned, cleared and stained

for bone (Alizarin Red)approx. 0.42–0.52L Organization of skin, SEM interconnections

between skin and myosepta6 [370 mm] Formalin fixed 0.18–0.25L Measurements of collagen fiber angles in skin

0.48–0.56L0.81–0.87L

7 [165 mm] Formalin fixed 0.17–0.24L Measurements of collagen fiberangles in skin0.54–061L

0.89–0.93L8 [210 mm] Formalin fixed 0.19–0.25L Measurements of collagen

fiber angles in skin0.49–0.53L0.85–0.89L

Body region is given relative to tip of snout (0.0L) and tip of caudal fin (1.0L). See text for further details.aFor histology all body specimens were embedded in paraffin following standard histological protocols. Section thickness was 10 lmin all cases. All sagittal and horizontal sections were stained according to the Azan-Domagk protocol (Romeis, 1986). In addition,some of the interserial sections were stained according to the Frankel protocol (Romeis, 1986). RM, red muscle; WM, white muscle.

CONNECTIVE TISSUE SYSTEM IN ANGUILLA ROSTRATA 31


bral segment). 3D images of the scanned region were generatedusing VG Studio Max. The lCT datasets from VG Studio Maxwere combined with the myosepta 3D data sets from Amira intoa single file using Cinema 4D software (.wrl- file format). Thisallowed us to combine in a single three-dimensional figure datafrom two sources, information on connective tissues from histol-ogy and information on skeletal elements from l-CT, and exam-ine the spatial relationships of the two types of tissues. Thecombined model was rotated until a perspective showing allthe relevant detail was obtained. Images of the skeletal and thecombined model were exported into Adobe Photoshop (.tif fileformat). Finally, the original unsmoothened myoseptum modelwas replaced by greytone artwork using Photoshop airbrushtools.

Investigation of the Skin and itsConnections to Myosepta

The skin and its connections to the myosepta were examinedhistologically (Table 1; Specimen 3) and by microdissection andScanning Electron Microscopy (SEM; Table 1, Specimen 5). Inthe latter case we used an unskinned cleared and stained speci-men from which we removed a square piece of skin �2.5 by 2.5cm with some of the underlying connective tissue (myosepta,endomysium, and stratum spongiosum) from the left epaxialside of the body immediately caudal to the anus, where thecross-section of the body was closest to being circular (see Fig.1). From this section, smaller specimens were excised. Skinspecimens underwent critical point drying and spattered withGold Palladium in preparation for SEM. The organizationof the skin and its interconnections with myosepta were cap-tured by a Cambridge Stereoscan 250 MK-2 scanning electronmicroscope.Skin-myoseptal interconnections were also examined histolog-

ically (Table 1; Specimen 3). Sections were taken in two obliquetransverse planes. One plane is parallel to the craniodorsal ori-entation of the collagen fibers in the skin, which is also the ori-entation of scale rows; therefore we term these sections parase-rial sections (see Gemballa and Bartsch, 2002). The other planeis parallel to the caudodorsal orientation of the collagen fibersin the skin, intersecting the scale rows and is therefore termedinterserial. Sections were examined under a microscope with

both transmitted light and polarized light (Zeiss Axioplan 2with polarized light device). The polarizer was used to highlightthe collagenous interconnections between skin and myosepta.

The stratum compactum fiber angles, defined as the angle ofthe serial fibers to the long axis of the body, were recorded fromSpecimens 6, 7, and 8 (Table 1) after carefully removing a rec-tangular piece of skin from two body positions. These positionswere at 0.17–0.25L (anterior) and 0.49–0.61L (midbody). Bothof these body regions have a nearly circular body cross-sectionwhich should allow us to interpret our morphological measure-ments with respect to models of organisms with circular cross-sections and fiber reinforcement in their skin arranged in twohelices of opposite handedness around their long axis (e.g.,Alexander, 1987). The rectangular pieces of skin were then pho-tographed under a Leica dissecting microscope using a Nikoncoolpix digital camera. Photographs were imported to ImageJ(ImageJ, U. S. National Institutes of Health, Bethesda, MD)and the fiber angles measured at five locations for each direc-tion, caudoventral and caudodorsal, using the remnants of thehorizontal septum attachment to the skin as a reference for thelong axis of the body.

Association of Myoseptal Tendons andRed Muscle Fibers

We used the sagittal, paraserial, and interserial sections (Ta-ble 1; Specimen 3) to detect specific associations of red musclefibers and myoseptal tendons. In these sections myoseptal ten-dons could easily be detected because they are thicker than anyof the remaining parts of myosepta (Gemballa et al., 2003a).Muscle fibers that insert into these tendons are either fastwhite muscle fibers or slow red muscle fibers. Both types offibers can be distinguished by their appearance (e.g., Videler,1993): red muscle fibers are slender and are oriented longitudi-nally whereas white muscle fibers are thicker and oriented atan angle to the long axis of the body. We paid particular atten-tion to the insertion of red muscle fibers into the tendons inorder to reveal putative pathways along which red muscleforces might be delivered during steady swimming.

RESULTS3D-Shape and Architecture of Myoseptaand Horizontal Septum

The terminology we use here to describe the dif-ferent parts of a myoseptum follows Gemballaet al. (2003a). Myosepta are bisected into an epax-ial and a hypaxial half at the midhorizontal levelby the horizontal septum and exhibit the typicalW-shape of most other fishes. The epaxial half ofthe myoseptum (Fig. 2A) bears an anteriorly-point-ing cone, the dorsal anterior cone (DAC), and aposteriorly-pointing cone, the dorsal posterior cone(DPC). The myoseptal part connecting these twocones is termed the epaxial sloping part (ESP).The dorsal most part that links the DPC to thedorsal midline is termed the epaxial flanking part(EFP). The hypaxial half is almost the mirrorimage of the epaxial half with correspondent cones(ventral anterior cone VAC, ventral posterior coneVPC) and correspondent myoseptal areas (hypaxialsloping part HSP, hypaxial flanking part HFP).

Three collagen fiber tracts are visible in each ofthe epaxial and hypaxial parts of the myoseptum(Figs. 2 and 3). One tract of fibers, the myorhab-doid tendon (MT), spans the entire flanking part of

Fig. 1. SEM specimen preparation. A flap of epaxial skinwas excised from a cleared and stained eel. The skin was pulledback to reveal the two main dermis layers: the stratum spongio-sum (red) with its irregularly oriented collagen fibers, and thestratum compactum (white) with its highly-oriented, crossedcollagen fibers. Myoseptal attachments (dotted line) can be seenon the medial side of the excised skin. White arrows indicatethe perspective of SEM images.

32 N. DANOS ET AL.


each myoseptum in an almost longitudinal orienta-tion from its anterior tip towards the posteriorcone. However, this tendon is only present in thelateral region of the flanking parts. The medialpart is devoid of longitudinal collagen fibers andremains thinner (see also section on ‘‘Insertions ofred muscles to myoseptal tendons’’).

The second and third distinct collagen fiber tractslie in the sloping parts (ESP and HSP). The lateraltendon (LT) connects the tips of the anterior and theposterior myoseptal cones, spanning the whole slop-ing part (ESP or HSP) of a myoseptum. As with theMT, this tendon is only present in the lateral regionof the myoseptum and thick longitudinal collagenfibers are absent from the medial part of the myo-septum. The epineural (ENT), or epipleural in thehypaxial region (EPT), tendon inserts into the verte-bral column and runs caudolaterally towards theskin in the sloping parts of the myoseptum. Itintersects the LT at an acute angle in the lateralpart of the myoseptum. Epineural and epipleural

tendons (ENT, EPT) are the only myoseptal tendonsthat insert to the vertebral column. Their insertionpoints, as revealed by microdissections and l-CTscans (Fig. 2B), lie on the neural arch in the case ofthe ENT and on the hemal arch in the case of theepipleural tendon. Immediately lateral to this inser-tion on the vertebral column, the tendons are ossi-fied along most of their way towards the skin(termed epineural and epipleural bones ENB, EPB;Fig. 2A).

The insertions of the ENT and EPT lie along themedial insertion line of each myoseptum along thevertebral column. This myoseptal insertion ontothe vertebral column spans three vertebral seg-ments (Fig. 2A). It starts at the anterior margin ofthe first vertebra and runs caudally across theneural arch of this vertebra. After passing thisneural arch (including the insertion of the ENT)the insertion line passes two more neural archesuntil it aligns with the neural spine of the thirdvertebra. Having reached the end of this spine the

Fig. 2. A: Graphic representation of the epaxial part of an eel myoseptum at 0.60L. 3D-reconstruction of skeletal elements is basedon a lCT scan, myoseptum was reconstructed from serial histological sections, and its attachment to axial structures was obtained frommicrodissections of cleared and stained specimens. The myoseptum is shown semitransparent. Gray lines indicate collagen fiber tracts.The lateral attachment line (lal) along which the myoseptum inserts into the skin is indicated as a white line enframed by fine darklines; the white dotted line marks the insertion of the myoseptum to the vertebral axis. The area dorsal to the DPC is called EFP; thisarea contains the MT. The area between dorsal anterior cone (DAC) and DPC is called ESP; this area contains the LT and the ENT. Notethat the proximal part of the ENT is ossified (ENB). ENT fibers diverge from this ENB. B: Detail of proximal end of ENB. The whitecircle shows the insertion of the ENB into the neural arch. fr, fin rays; hs, horizontal septum; pt, pterygiophores; v, vertebral column; vs,vertical septum.



insertion line takes a turn in a craniodorsal direc-tion at which it remains until its end at the dorsalmidline (Fig. 2A).

The described features of myosepta do notchange along a rostrocaudal gradient in the speci-

mens examined. Myoseptal shape was similarthroughout the trunk and the tendons described,including their ossifications, appeared similar in allbody regions. Moreover, the overall length of myo-septa (i.e., the rostrocaudal span reaching from thetip of the anterior to the tip of the posterior myo-septal cone), a feature used to compare myoseptaalong a rostrocaudal gradient, remains constantalong the whole myoseptal series. The overall myo-septal length equals the span of the LT parallel tothe long axis of the body from the tip of the ante-rior to the tip of the posterior cones. We obtainedoverall myoseptal length values between 4.4% and5.1% of body length (Table 2).

The horizontal septum consists of a system ofcrossing collagenous fibers (Fig. 4A). One set offibers forms distinct tendon-like structures, theepicentral tendons, which run caudolaterally fromtheir insertion at the vertebral column to the skin.Epaxial and hypaxial myosepta insert onto thehorizontal septum along an epicentral tendon. Thesecond direction of fibers is represented by evenlydistributed caudomedial fibers that are termedposterior oblique fibers. In the posterior body only,these fibers are more prominent in the medial partof the horizontal septum and might be bettertermed here posterior oblique tendons (POTs).However, these POTs are only present close to thevertebral column and are absent in the lateralpart of the horizontal septum (Fig. 4B).

The Dermal Connective Tissue System andits Connections to the Myosepta

Organization of the dermis. Using SEM wewere able to visualize the two thickest layers ofthe dermis, the stratum compactum with its highlyordered collagen fibers and the stratum spongio-sum with its randomly oriented fibers (see Fig. 5).The stratum compactum from the lateral midbodysection of the specimen photographed (TL 5 25cm) comprises 30 collagen lamellae, with eachlamella made up of only one layer of parallel colla-gen fibers (Fig. 5A). The histological micrographsshow only those lamellae that have parallel fiberorientations to the sectioning orientation and themean number of visible lamellae was 13.58, ascounted in nine micrographs from the midbodyregion (s.d. 5 1.33). This number is approximatelyhalf of the number measured in the SEM sectionin which lamellae with fibers in both directionscan be distinguished, suggesting that histologicalpreparation artifacts allow only fibers parallel tothe plane of section to be clearly viewed.

The mean stratum compactum collagen fiberangles, the angle at which the collagen fiberswithin a lamella run parallel to the long axis ofthe body, were 57.58 6 6.928 for the caudodorsaldirection (n 5 30) and 60.51 6 7.058 for the caudo-ventral direction (n 5 30). Of the three specimens

Fig. 3. Eel myoseptum of posterior body region (0.75L)spread out under polarized light. Tendinous fiber tracts are visi-ble as white strands. The membranous ossification (ENB, seeFig. 2) cannot be distinguished from the collagenous fiber tractof the ENT. Note that epaxial and hypaxial parts are symmetri-cal around the horizontal septum (gray dashed line). Prefix e-stands for epaxial, h- for hypaxial; EPT, epipleural tendon; seeFigure 2 for further legends. The medial attachment of themyoseptum to the vertebral axis and vertical septum is outlinedwith a white dotted line. Mediolateral from left to right, dorso-ventral from top to bottom.

34 N. DANOS ET AL.


measured the angles at anterior and midbody dif-fered neither between body regions nor in the cau-doventral and caudodorsal directions (Student’s ttest, a 5 0.05). In contrast, the collagen fibers inthe stratum spongiosum are arranged irregularlybut densely, forming a collagenous sheath (Fig.5B,C). Observations from gross dissections of 10individuals suggest that the stratum spongiosumvaries in thickness between individuals and at dif-ferent body locations.

Myosepta-skin interconnections. For thesample depicted in Figure 5B, the stratum spon-giosum was peeled away from half of the areabetween two adjacent myosepta, cross-sectionedparallel to the direction of the myoseptum andlifted up. This exposed the attachments of themyoseptum to the stratum compactum. We verifiedthat the observed attachments were of the myosep-tum and not of the stratum spongiosum, by pullingon myoseptal fibers and noticing that the attach-ment fibers following the same trajectory cameunder tension.

Figures 5C–G depict a section cut along thebase of two myosepta, from the horizontal septum

to the dorsal anterior-pointing cone. The myosep-tum was cut close to its attachment with theskin, hence the free fiber ends in Figure 5B,C.The stratum spongiosum was peeled back fromone myoseptum toward the other to expose itsattachments to the stratum compactum. Themyosepta connect to the stratum compactum viadiscrete fiber bundles that run all the way fromthe myoseptum through the stratum spongiosumto the stratum compactum. Both SEM and histo-logical investigation confirmed that two kinds ofconnections are present: transverse fiber bundles(tfb; Fig. 5D,F,G) or weaving fiber bundles (wfb;Fig. 5E,F). The transverse fiber bundles cuttransversely through all the stratum compactumlayers and insert in the basal lamella, the layerbetween the dermis and the epidermis. Trans-verse fiber bundles are more often observeddirectly lateral from the myoseptum (Fig. 6B,C).The weaving fiber bundles on the other handbranch anteriorly and posteriorly from the myo-septal-stratum spongiosum interface and weaveinto stratum compactum collagen fibers at multi-ple lamellae, beginning with the medial-most

TABLE 2. Rostrocaudal span (given as % of body length) of epaxial and hypaxial lateral tendonsat three axial positions (0.4L, 0.6L, and 0.75L) in three specimens of Anguilla rostrata

Axial positionSpecimen 1[290 mm TL]

Specimen 2[274 mm TL]

Specimen 3[295 mm TL] Mean

0.40L (epaxial) 5.05% 4.88% 4.66% (at 0.45L) 4.86% 6 0.16a

0.40L (hypaxial) 3.42% 3.49% 4.05% (at 0.45L)0.60L (epaxial) 4.98% 4.81% 4.85% 4.85% 6 0.160.60L (hypaxial) 4.65% 5.09% 4.71%0.75L (epaxial) 4.41% 4.47% 4.58% 4.54% 6 0.090.75L (hypaxial) 4.51% 4.64% 4.62%

aHypaxial myosepta from the anterior body generally differ in shape (e.g., less pronounced cones andtendons; e.g., Gemballa and Roder, 2004) from myosepta of other body regions due to the presence ofthe outbulging peritoneal cavity. Hence, only epaxial myosepta were used for the calculation of themean value at 0.40L.

Fig. 4. Polarized light micrographs of the horizontal septum. The crossing fiber directions of the epicentral tendon (ECT) andthe posterior oblique fibers/tendons (POF/POT) are indicated by the arrows. A: Midbody region, top view of left body side. Medialinsertion to vertebral column on top, cranial to the left. B: Posterior body region, top view of left body side. Medial insertion to ver-tebral column on top, cranial to the left. The horizontal septum is cut along the anteriormost epicentral tendon (white dotted line).The collagenous fibers to the left of this dotted line belong to the myoseptum (MS) that inserts into the horizontal septum. Notethat fibers of two directions (asterisk and double-asterisk) run from the myoseptum into the horizontal septum.



Fig. 5. SEM images of Anguilla rostrata skin. A: Cross-section of the stratum compactum, illustrating the plywood arrangementof alternating collagen fiber layers at approximately 908 to each other. Section plane is parallel to one fiber orientation and perpen-dicular to the other. B: Mediolateral view of skin, showing the attachment of an epaxial myoseptum (ms) to the skin. The stratumspongiosum (ss), which separates the muscle fibers from the skin has been pulled back, cross sectioned and lifted up to expose theconnections of the myoseptum with the stratum compactum. The irregular arrangement of collagen fibers in the stratum spongio-sum can be seen in contrast to the regular arrangement in alternating fiber layers on the medial surface of the stratum compactum(sc). C: The stratum spongiosum (ss) is partially removed to expose its connections to the stratum compactum. D: Magnification ofthe white square in (C). The two kinds of connections attaching the stratum spongiosum and the myoseptum to the stratum com-pactum, as seen under a lifted subdermal layer. E: Magnification of the white square labeled ‘‘E’’ in (D). The weaving fiber bundleextends from the stratum spongiosum and joins the weave of the stratum compactum. Fibers from the bundle run in both stratumcompactum directions and dive into more lateral layers of the stratum compactum. F: Magnification of the white square labeled‘‘F’’ in (D). The transverse fiber bundle originates more medially in the stratum spongiosum. No apparent fibers from this bundlefollow the orientation of either stratum compactum layer but instead cross the stratum compactum perpendicularly. G: Ultrastruc-ture of a transverse fiber bundle. Collagen fibers dive into the stratum compactum. Each fiber consists of smaller fibrils. Fat glob-ules can be distinguished on the stratum compactum. Ms, myoseptum; sc, stratum compactum; ss, stratum spongiosum; tfb, trans-verse fiber bundle; wfb, weaving fiber bundle.

36 N. DANOS ET AL.


lamella. These branches from the stratum spon-giosum to the stratum compactum are neverobserved to cross the branches from an adjacentmyoseptum, suggesting that there are regions ofthe myomere that are not directly attached to theskin (Fig. 6A).

A third type of connection is obvious from somehigh-resolution histological micrographs (Fig. 6B,yellow arrow). This type connects two adjacentstratum compactum layers and the distribution ofthis type of connections is staggered and evenlydisbursed throughout the stratum compactum.

The Musculotendinous System: Insertionsof Red Muscles to Myoseptal Tendons

Red muscles in eel form a superficial layer thatextends far epaxially and hypaxially (Fig. 7A) andonly extends medially towards the vertebral col-umn along the horizontal septum (‘‘deeper’’ redmuscle). In these regions it can be as deep as onethird to one half of the distance from the skin tothe vertebral column. Red muscle not only coversthe two inner legs of the ‘‘W’’ shape of myosepta(i.e., the epaxial and HSP) but also part of the dor-sal- and ventral-most flanking parts. For mostparts of its dorsoventral extension the red musclecover remains a thin layer. The thickness of thislayer is 55.8 6 4.7 lm (n 5 13) in the flankingparts and 89.7 6 13.6 lm (n 5 17) in the slopingparts.

We describe the spatial associations of red mus-cle fibers, myosepta, and the horizontal septumfrom the histological sections (horizontal, sagittal,para- and interserial sections; see Table 1). Redmuscle fibers insert into the lateral-most parts ofepaxial and hypaxial myosepta in the sloping andflanking parts; they do not insert into the horizo-nal septum. However, the ‘‘deeper’’ red musclefibers that run along the horizontal septum insertinto the myoseptum which itself attaches to thehorizontal septum. Therefore, in order to bestdescribe red muscle insertions into myosepta andthe red muscle-tendon associations we need toaddress two aspects: insertions of red muscle fibersto myoseptal tendons of the lateral-most flankingand sloping parts, and insertions of red musclefibers to those myoseptal parts that are close tothe horizontal septum.

The lateral-most area of the flanking parts isformed by the MT. In histological sections, wedetected these tendons by their thickness (40.4 612.4 lm; n 5 17) that is significantly differentfrom more medial areas of the flanking parts (8.96 3.8 lm; n 5 13, see also Fig. 7B). Red musclesinsert into these robust MT (Fig. 7B). This muscle-tendon association is present in the epaxial as wellas in the HFP.

As in the flanking parts, myoseptal tendonsalso form the lateral-most area of the myoseptal

Fig. 6. Micrographs of histological specimens. A: Myoseptumat the ESP level inserting into the stratum compactum. Thestratum compactum is comparable in thickness to the epider-mis. Two thick fiber bundles, stained blue, extend from themyoseptum through the stratum spongiosum. One fiber bundleinserts into the stratum compactum in this section via a weav-ing fiber bundle (left fiber) while the fiber bundle on the rightlooks like a transverse fiber bundle. B: Higher resolution ofmyoseptum-skin connection from a HSP region. There is anexample of a weaving fiber bundle and an example of a trans-verse fiber bundle. Yellow arrow points to the interlamellar con-nections. C: Same view as 6B but under polarized light. Parallelcollagen fibers appear in bright yellow. All micrographs are ofcross sections along one stratum compactum fiber direction. Ep,epidermis; ms, myoseptum; rm, red muscle; sc, stratum compac-tum; scl, scale; ss, stratum spongiosum; tfb, transverse fiberbundle; wfb, weaving fiber bundle.



sloping part. These LT were visible as relativelythick structures of connective tissue in histologi-cal sections (34.2 6 11.5 lm; n 5 12, as opposedto thin non-tendinous regions with a mean thick-ness of 9.9 6 4.4 lm; n 5 11, see also Fig. 7C). Inthis region too, red muscle fibers insert into theserobust LT.

Sagittal sections show that red muscle fibersare not directly connected to the horizontal sep-tum. Rather, these muscle fibers run longitudi-nally, connecting adjacent myosepta (Fig. 7D)and inserting into collagenous myoseptal fibersthat run caudolaterally. The myosepta in turninsert into the horizontal septum. The myoseptalfibers to which red muscle fibers attach are at

obtuse angles with the posterior oblique tendon(POT) fibers of the horizontal septum (Fig. 4B).Therefore, any transmission of red muscle forcesto the POT would be directed along the path ofthe POT to the vertebral column. The maximumdistance across which any force transmission viathe horizontal septum can occur is the distancespanned by the POT fibers. This span was deter-mined to be around 1.2%L (1.23 6 0.12%L; n 5 9measurements in Specimens 1 and 2). In contrastto red muscle fibers, the more medially placedwhite muscle fibers insert directly into collage-nous myoseptal fibers on the horizontal septumparallel to POTs that are found in the septum(Fig. 4B).

Fig. 7. Arrangement of red muscles in Anguilla rostrata and insertions of red muscle fibers to specific myoseptal tendons. A:Overview of the thin layer of red muscle (encircled by dashed line) in the left epaxial midbody region; paraserial section. Note thatthe superficial thin layer of red muscles is well separated from the deeper white muscles. Along the horizontal septum red musclesenter deeper regions of the trunk. B: Insertion of red muscles into MT in EFP. Paraserial sections of left body side. Note that thered muscle fibers and lateralmost white muscle fibers insert into the thick tendon whereas more medial white muscle fibers insertinto the thin part of the myoseptum. C: Polarized light micrographs showing two examples of insertions of red muscles into LT inHSP. Paraserial sections of left body side. D: Insertion of red muscles into myoseptal part close to horizontal septum. Note that thered muscle fibers do not insert into the horizontal septum. Sagittal section of left body side; cranial to left. Ep, epidermis; hs, hori-zontal septum; ms, myoseptum; rm, red muscles; sc, stratum compactum; v, vertebral axis; vs, vertical septum.

38 N. DANOS ET AL.


DISCUSSION3D-Shape and Architecture of Myosepta andHorizontal Septum

Comparison of myoseptal architecture:Eels versus non-anguilliform swimmers. Thisstudy is the first to describe the architecture,3D-morphology, and vertebral insertions of myo-septa in an anguilliform swimmer. Our data showthe presence of one anteriorly- and one posteriorly-pointing cone and a set of three myoseptal tendonsin each the hypaxial and epaxial parts of the myo-septum of Anguilla rostrata (Figs. 2 and 3). Boththe cones and the three tendons (myorhabdoid, lat-eral and epineural in the hypaxial region or epi-pleural in the epaxial tendon) have been shown tobe present in a number of teleost and non-teleostfishes, including basal actinopterygians and repre-sentatives of all major clades of chondrichthyans(Gemballa et al., 2003a; Gemballa and Hagen,2004; Gemballa and Roder, 2004). Except for theMT which has been identified as an apomorphicfeature of the Myopterygii (Petromyzontiformes 1Gnathostomata) these myoseptal features repre-sent apomorphic features of the Gnathostomata(Gemballa et al., 2003a, 2006). All of the abovemyoseptal features have been retained in theAmerican Eel (Figs. 2 and 3). The same is true forthe insertion of the myoseptum to the vertebralcolumn which in the eel as well as in other actino-pterygian fishes spans three subsequent vertebrae(Fig. 2 of this study; Gemballa et al., 2003a).

In Anguilla rostrata, as in other teleosts, most ofthe ENT and EPT tendons are partly ossified,forming the ENB and EPB (see Fig. 2). In teleostssuch membranous ossifications have been observedin all myoseptal tendons to various degrees.Though they generally follow a rostrocaudal pat-tern of ossification, with the rostralmost tendonsbeing ossified more often, the degree of ossificationvaries largely even between closely related species(Patterson and Johnson, 1995; Gemballa et al.,2003a). Since there is currently no experimentalevidence (e.g., dynamic bending experiments of thestructure) pointing to the function of these rod-likeossifications and since their systematic distribu-tion is irregular, we do not at this time propose afunctional explanation for the existence of suchintermuscular bones.

Recent morphological studies have revealedexceptions to the gnathostome groundpattern ofmyoseptal architecture in fishes with differingswimming modes. Carangiform and thunniformswimmers lack a distinct ENT and EPT. Further-more, in these species the length of the LT (i.e.,the distance spanned between the anterior andposterior cones) increases twofold to threefold fromtendons of the anterior body region to the posteriorbody region (Gemballa and Treiber, 2003; Gem-balla, 2005; Shadwick and Gemballa, 2006; Gem-balla et al., 2007). The posterior LTs in carangi-

form and thunniform swimmers usually span 0.2L.Both of these features, the lack of ENT and EPTand the elongation of LTs, have been noted in car-angiform and thunniform swimmers from a widetaxonomic range (e.g., lamnid sharks, carangids,swordfish, basal scombrids, and tunas) but havenever been reported for any subcarangiformswimmer (for review see Shadwick and Gemballa,2006).

In contrast, LTs of subcarangiform swimmersare short and relatively uniform along the body.For example, tendon lengths lie between 0.050Lanteriorly and 0.056L posteriorly in catsharks, andbetween 0.072L anteriorly and 0.082L posteriorlyin the perciform Channa (for corresponding datain 11 species see Shadwick and Gemballa, 2006;average for subcarangiform swimmers is 6.4 6 0.9anteriorly and 7.5 6 1.1 posteriorly) and between0.086L anteriorly and 0.094L posteriorly in theperciform Lepomis gibbosus (Gemballa, unpub-lished data). A similar LT average length androstrocaudal length uniformity has also been dem-onstrated for the American Eel in this study(around 0.045L to 0.05L anteriorly and posteriorly;Table 2). Thus, with respect to this first exampleanguilliform swimmers are morphologically similarto subcarangiform swimmers but differ consider-ably from carangifom and thunniform swimmers.We discuss the functional significance of this pat-tern in the light of muscle dynamics and swim-ming kinematics in the section titled ‘‘Characteri-zation of Undulatory Swimming Modes.’’

Comparison of the horizontal septum: Eelsversus non-anguilliform swimmers. We founda cross-fiber array in the horizontal septum ofAnguilla rostrata (see Fig. 4) that is similar tothat of other actinopterygians. Both fiber direc-tions are present in most species studied so far(�35 actinopterygians; Gemballa et al., 2003b).The epicentral tendons show little variation amongspecies, whereas fibers of the second direction maybe evenly distributed (i.e., posterior oblique fibers;e.g., in the anterior body region of Anguilla) orform distinct tendons (i.e., POTs; e.g., in the poste-rior body region of Anguilla). However, this varia-tion does not correlate with swimming modes.Independent of the swimming type, a species mayhave POTs or posterior oblique fibers or even lackthe whole horizontal septum all together.

The Skin Connective Tissue System and itsConnections to the Myosepta

Organization of the skin. Our SEM images ofthe stratum compactum (Fig. 5) show that lamellaeare composed of only one layer of collagen fibers, inaccord with the results from a previous histologicalstudy of Anguilla rostrata skin (Leonard andSummers, 1976) but in contrast to a study ofLepadichthys lineatus skin (Fishelson, 1972) and a



developmental study in Danio rerio (Le Guellecet al., 2004). The presence of more than one layer ofcollagen fibers per lamella seems to be the most com-mon condition among other vertebrates with a simi-larly ordered stratum compactum (Olsson, 1961;Fujii, 1968; Fishelson, 1972; Hawkes, 1974; Mittaland Banerjee, 1974). We counted 30 lamellae fromthe lateral side of the body from one SEM imageand 13.58 6 1.33 parallel lamellae from nine micro-graphs. Other studies of fish integument have found15 to 30 lamellae in the goby Chasmichthys gulosus(Fujii, 1968), between 8 and 12 in seven species ofsharks (Motta, 1977), 15 to18 in Neoceratodus and17 to 25 in polypterids (Gemballa and Bartsch,2002). American eels and the goby, C. gulosus, havethe highest number of lamellae of the species exam-ined, suggesting a thick and tough skin. Both ofthese species experience high mechanical abrasionwhen locomoting in contact with the substrate inshallow freshwater or tide pools or even on land. Atougher but smooth skin with diminutive scales em-bedded in the epidermis as in the American eel mayprovide mechanical protection and reduced frictionagainst the substrate (Fishelson, 1996).

Fiber angle, the angle at which a fiber lies withrespect to the long axis of the body, has been usedto predict the mechanical effects on bending of ply-wood-like structures that wrap around animalbodies in a helical pattern. Alexander (1987) identi-fies an angle of 608 as the angle at which extensiblefibers in the body walls of helically wound animalsdo not change length during bending. Under thesame model, fiber angles less or more than 608 tendto increase the elastic strain energy of the systemin the bent position, causing the body to springback to the straight position. Harris and Crofton(1957) calculate that a cylinder has maximum vol-ume when enclosed by a plywood-like arrangementwith a fiber angle of 558. Cylinders under torsiontend to break at 458 to their long axis and Hebrank(1982) who measured fiber angles in Anguilla ros-trata indirectly by measuring the angle betweenscale rows and the long axis of the body found themto be approximately 458.

The 95% confidence intervals of the fiber anglesmeasured in this study, 57.58 6 6.928 for the cau-dodorsal direction and 60.51 6 7.058 for the caudo-ventral direction, contain all the above criticalangles. This suggests that there is the potentialfor elastic storage in the skin of the American eel,but we cannot confirm that this is indeed occur-ring during locomotion. Although our results differfrom those of Hebrank (1982), we still consider re-sistance to torsion to be a major function of dermalcollagen. Collagen is a strong tensile material(Wainwright et al., 1982) and any tensile forceapplied to the fiber, such as torsion around thevertebral column as experienced by the skin, willbe resisted. Any torsional forces generated duringswimming will be maximal at the farthest distance

from the vertebral column and at an angle close to458 with the longitudinal axis of the body (Vogel,2004). This is the angle at which we would expectfiber reinforcement of the body walls if we modelthe eel body as a cylinder with circular symmetryin its material properties as well as symmetricalmotions generating torsion. However, we knowthat the cross-section of an eel is neither perfectlycircular nor symmetrical in its material properties.Furthermore, the presence of fins and the kine-matics of eel swimming (Tytell, 2004a,b) are fac-tors that introduce asymmetry to the pattern oftorsional forces experienced by the body. With across-section that is oblong in the dorsoventraldirection we are not surprised to find that fibersare directed more in that direction. The overalldermal collagen fiber arrangement allows for lat-eral flexibility of the body while resisting torsionalforces.

The subdermis, or hypodermis as it is also some-times called, contains the stratum spongiosum, asheath of irregularly arranged collagen fibers andlipid storage. The specimens examined, all in themigratory elver stage, had ample fat storages.Other researchers have noticed large fat stores inthe air-breathing Channa striata and suggestedthat the fat may function as a water repellant pre-venting water loss when under desiccation stressor as a shock-absorber when C. striata buries inthe mud (Mittal and Banerjee, 1975a). Eventhough eels regularly locomote on land or in shal-low marshy fields, given the catadromous migra-tory stage the animals in this study were in, it ismore likely that the fat in the stratum spongiosumwas for energy storage.

Myosepta-skin interconnections. Verticalbundles of collagen fibers, originating in the myo-septa, traversing the stratum spongiosum and join-ing fibers of the stratum compactum have beenobserved at varying densities in nearly all previoushistological studies of fish skin including sarcop-terygian fish as well as in studies of Amphioxus,anuran and snake skin (Chapman and Dawson,1961; Olsson, 1961; Liem, 1967; Logan and Odense,1974; Jayne, 1988; Gemballa and Bartsch, 2002).These bundles terminate in the basal membrane,the layer between the stratum compactum and theepidermis, and do not pass into the epidermis(Mittal and Banerjee, 1975b; Leonard andSummers, 1976). The bundles provide bondingstrength to the stratum compactum lamellae andadd strength to the whole plywood configuration.The three types of perpendicular fibers describedearlier (Figs. 5E,F and 6B), have also beendescribed in detailed morphological studies of per-pendicular fiber bundles in the skin, such as that ofAmphioxus (Olsson, 1961) and of Danio skin (LeGuellec et al., 2004). The perpendicular bundleswere not present in a leptocephalus larva examinedin one study (Leonard and Summers, 1976).

40 N. DANOS ET AL.


Evaluation of Possible Ways of Red MuscleForce Transmission

There are currently no techniques to directly orindirectly measure forces along tendons or tissuesin swimming fishes at such small scales. Thus, ourunderstanding of force transmission in swimmingfishes is mainly inferred from morphology. Threeconnective tissue structures have been hypothe-sized to play a role in force transmission in fishes:myoseptal tendons, POTs of the horizontal septum,and the stratum compactum of the skin (Wain-wright et al., 1978; Hebrank, 1980; Hebrank andHebrank, 1986; Westneat and Wainwright, 2001;Gemballa and Treiber, 2003; Donley et al., 2004;Gemballa et al., 2006). In the following sections weevaluate whether our results in the American eelare consistent with these hypotheses.

Direct observations of insertions of red musclefibers into myoseptal tendons in some species sug-gest that these tendons play a role in force trans-mission (Gemballa and Treiber, 2003; Donleyet al., 2004; Gemballa et al., 2006). The signifi-cance of myoseptal tendons as force transmitterswas further supported by a computational modelthat suggests tendons might act to transmit forcesand add body flexural stiffness, and by findings ina shark species in which the length of the LTagreed with the distance of force transmission pre-dicted from muscle dynamics and swimming kine-matics (Long et al., 2002; Donley et al., 2004, 2005).

In this study we show that in Anguilla rostratared muscles insert into the lateral and myorhab-doid tendons (Fig. 7B,C). Thus, red muscle forcesmight be transmitted posteriorly through thesetendons. These tendons, however, are remarkablyshort (around 0.045–0.05L; see Table 2) and thered muscle fibers do not insert at the anterior endsbut in the middle of the tendons. Hence, thisarrangement suggests that forces will be trans-ferred only a short distance (maximum of 0.025L).

This insertion of red muscle fibers in the midregion of a LT has also been identified in subcar-angiform and carangiform swimmers (Gemballaand Treiber, 2003; Gemballa et al., 2006; Shadwickand Gemballa, 2006). Since subcarangiform andanguilliform swimmers are also similar in LTlengths (see above section titled ‘‘Comparison ofMyoseptal Architecture: Eels vs. Non-AnguilliformSwimmers’’) they appear to share a similar muscu-lotendinous design. This overall morphologicalsimilarity is consistent with the view that the mid-line kinematics in these two swimming types ismore similar than previously thought (Lauder andTytell, 2006).

The musculotendinous system of carangiform andthunniform swimmers differs from that of anguilli-form or subcarangiform swimmers in some ways.First, carangiform and thunniform swimmers bearelongated LTs (see earlier section titled ‘‘Comparison

of Myoseptal Architecture: Eels vs. Non-Anguilli-form Swimmers’’). Second, in thunniform swimmersred muscle inserts into the anterior part of theelongated LT. Thus, red muscle forces will be trans-mitted over a long distance. Such red muscle-ten-don arrangements have been identified in tunasand lamnid sharks and for both groups and it hasbeen shown experimentally that bending occursmore caudally than muscle contraction (Shadwicket al., 1999; Katz et al., 2001; Donley et al., 2004,2005; Gemballa, 2005). In contrast, as a conse-quence of force transmission over a short distancecaudally in non-thunniform species, we expectmuscle contraction to lead to body bending close tothe site of and in phase with muscle contraction.Indeed, this has been demonstrated for severalsubcarangiform and carangiform swimming tele-osts and one shark (Coughlin et al., 1996; Shad-wick et al., 1998; Katz et al., 1999; Donley andShadwick, 2003). On the basis of our morphologicaldata we predict that this will also hold true foranguilliform swimmers.

Although the horizontal septum has beenhypothesized to be part of the red muscle forcetransmission system in scombrid fishes (Westneatand Wainwright, 2001), we cannot extend this hy-pothesis to include the American eel. First, redmuscles that are placed close to the horizontal sep-tum (i.e., ventrally to the anterior cone) do notinsert into the horizontal septum but into the myo-septum (Fig. 7D). As these muscle fibers contractthey will pull on the myoseptum but these forcesare unlikely to be transferred into the horizontalseptum because the collagenous fibers in the myo-septum and horizontal septum are at obtuseangles that are not efficient for force transmissionfrom the myosepta to the horizontal septum (Fig.4B). In contrast, medially placed white musclefibers will pull on myoseptal fibers that are in linewith the POTs of the horizontal septum (POTs;Fig. 4B). Thus, muscular forces might be trans-ferred via POTs to the vertebral column. This forcetransfer, however, will only span a short distance(around 0.01L; Fig. 4B).

The idea that the stratum compactum is part ofthe force transmission system in fishes was firstproposed for the lemon shark, Negaprion brevirost-ris (Wainwright et al., 1978). In that study theauthors identified a fiber angle at which the skinwould become extremely stiff and thus act as atendon. This angle was also identified in the Amer-ican eel and it was inferred that eel skin wouldalso act as an exotendon (Hebrank, 1980, 1982).These studies assumed that internal body pressurewas putting the skin under tension but severalstudies have found that there are different inter-nal pressures on contralateral sides of the body,negating the simple transformation of internalbody pressure into body wall stress by modelingthe body as thin-walled cylinder (Westneat et al.,



1998; Danos, 2005). Therefore, the only other waythat the skin can be put under tension and thusact as a force transmitter would be by direct physi-cal connection of the force-producing elements, themuscle fibers, to the force transmission tissue, themyosepta and skin. But the results of our presentstudy demonstrate that neither red nor whitemuscles directly insert into the skin. On the basisof our morphological findings presented here, theonly way of transmitting muscular forces to theskin would be through the transverse or weavingfiber bundles (tfb, wfb; Figs. 5 and 6). These con-nections, though, do not appear very robust whencompared with the LTs or MT (Figs. 6 and 7).Moreover, they do not form continuous intercon-nections between myoseptal tendons and skin butare present as single fiber bundles that are evenlydistributed along a myoseptum (Fig. 5B,C).

Studies of body stiffness at different phases rela-tive to the strain cycle of the muscles in electri-cally stimulated eels suggested that net positivepower in the posterior muscle of eels is, in part,produced by an elastic strain energy mechanism(Long, 1998). At that point too little was knownabout the musculotendinous system in eels to findthe morphological correlate for the serial elasticelements that were hypothesized to account forelastic energy storage in eels. From the morpholog-ical findings presented here we hypothesize thatthe lateral and MT might contribute substantiallyto the body’s mechanical properties. However, asthe predictions derived from the analysis of bodystiffness only partly match the results of in vivostudies of muscle activity (Gillis, 1998a), the ideaof an elastic strain energy mechanism remainsequivocal.

Characterization of UndulatorySwimming Modes

In the classical review of fish swimming modes,Lindsey (1978) distinguished three main types ofaxial undulatory swimming: anguilliform, sub-carangiform, and carangiform. These swimmingmodes were thought to form a continuum with thesubcarangiform mode representing an intermedi-ate mode. The differentiation of these swimmingtypes is based on two parameters that are derivedfrom 2D-midline kinematics, the length-specificbody amplitude as a function of body position (A/L), and length-specific wavelength of the propul-sive wave (k/L). However, recent comparative stud-ies revealed that the midline kinematics ofthe anguilliform mode is unexpectedly similar toother modes, especially the subcarangiform mode(Lauder and Tytell, 2006). During slow swimming(<2 L/s) the specific tail beat amplitude liesbetween 0.08 and 0.12L in anguilliform swimmersand between 0.07 and 0.13L in subcarangiformswimmers (Videler, 1993; Gillis, 1997, 1998b). Fur-

thermore, amplitudes in the whole posterior halfof the body appear to be similar in all swimmingtypes (Lauder and Tytell, 2006).

If the parameters from traditional 2D-midlinekinematics do not adequately discriminate bet-ween undulatory swimming types, parametersfrom other fields are needed to either support thetraditional classification or redefine new types. Weconsider the morphology of the locomotory system,muscle dynamics, and hydrodynamics to be inte-gral features that affect swimming styles of fishes.Since recent studies, including this study on theAmerican Eel, have provided comparative data ineach of these fields we are able to update the cur-rent discussion on classification of swimmingmodes in fishes.

Anguilliform and subcarangiform swimmers aremorphologically very similar (eels: this study; tele-osts: e.g., Gemballa and Treiber, 2003; Gemballaet al., 2003a; Gemballa and Roder, 2004; Gemballaet al., 2006; Shadwick and Gemballa, 2006). Theyshare the same 3D-shape of myosepta and thesame set of myoseptal tendons. This arrangementdiffers substantially from carangiform and thunni-form swimmers that lack ENT and have elongateLTs in posterior body (see section titled: ‘‘Compari-son of Myoseptal Architecture: Eels vs. Non-Anguilliform Swimmers’’; Shadwick and Gemballa,2006). Thus, swimming types are well divided intocarangiform and non-carangiform by morphologi-cal features.

Recent comparative hydrodynamic studies haveidentified substantial differences in the wake pat-terns of eels and other teleosts that can be used todivide swimming types into anguilliform and non-anguilliform types (e.g, Tytell and Lauder, 2004;Lauder, 2006; Lauder and Tytell, 2006). Thesedifferences in wake patterns are thought to occurmainly due to differences in body shape. Non-anguilliform swimmers (especially carangiformswimmers) exhibit a marked narrowing of the lon-gitudinal shape anterior to the tail that is notpresent in anguilliform swimmers.

The length-specific propulsive wavelength (k/L)is one of the classical kinematic parameters usedby Lindsey (1978) that still seems to be valid fordefining undulatory swimming types. Recentstudies confirmed the classical view that k/L grad-ually increases from anguilliform to carangiformswimmers (0.6 to >1.2; e.g., Videler, 1993; Wardleet al., 1995; Donley and Dickson, 2000; Donley andShadwick, 2003; Dowis et al., 2003; Tytell andLauder, 2004). Furthermore, the length-specificpropulsive wavelength is correlated with red mus-cle activation patterns. Fishes that swim withgreat k/L have long burst durations (i.e., the timea certain muscle segment is turned on within atail-beat cycle) anteriorly and a large decrease(17–19%) in burst duration between anterior andposterior segments (review: Gillis, 1998a). As a

42 N. DANOS ET AL.


consequence, in carangiform swimmers anteriorsites still remain active (long-duration anteriorbursts) while the wave of activity reaches posteriorsites. Thus, muscle activity in carangiformswimmers is characterized by a long block of ipsi-lateral muscle activity. In contrast, anguilliformswimmers have relatively short burst durationsanteriorly with almost no decrease (0–5%) towardsposterior sites (Gillis, 1998a). Subcarangiformswimmers are at intermediate values (10–14%).Interestingly, the long block of ipsilateral muscleactivity in carangiform swimmers is associatedwith long myoseptal tendons in the posterior body(e.g., tendon lengths of 0.16–0.20L; Shadwick andGemballa, 2006) whereas in anguilliform and sub-carangiform swimmers short blocks of ipsilateralmuscle activity are associated with short tendonsin all body regions (e.g., around 0.05L in eels, thisstudy; 0.005–0.009L in other fishes, Shadwick andGemballa, 2006). However, a causal relationshipthat explains how these differences in musculoten-dinous design and muscle function are translatedinto different swimming modes (e.g., different pro-pulsive wavelengths) has not yet been established.

These examples from morphology, hydrodynam-ics, and muscle physiology demonstrate that differ-ences across the classical swimming types can befound when parameters like wake patterns, pro-pulsive wavelengths, red muscle activity patterns,and LT lengths are compared. Thus, the tradi-tional classification still appears to be justifiedalthough the various swimming types are verysimilar in 2D-midline kinematics. However, datafrom the various fields mentioned earlier, except inthe case of eels, were usually not collected on thesame species and thus the picture drawn may rep-resent a trend rather than a clear picture. Aclearer redefinition of swimming types will needfurther support from comparative studies thatobtain all relevant data from multiple species.

ACKNOWLEDGMENTS

We would like to thank E.L. Brainerd for hergenerous guidance and support. Our work bene-fited from the fine technical assistance of MonikaMeinert (Tubingen) for histology and CharlieHellmer for SEM imaging (Tubingen). 3D-recon-structions using Amira were done under theinstruction of Thomas Schmelzle (Tubingen). Themanuscript benefited from comments by M. Deanand E.M. Standen on earlier versions.

LITERATURE CITED

Alexander RMcN. 1969. The orientation of muscle fibers in themyomeres of fishes. J Mar Biol Assoc UK 49:263–290.

Alexander RM. 1987. Bending of cylindrical animals with heli-cal fibres in their skin or cuticle. J Theor Biol 124:97–110.

Brainerd EL, Patek SN. 1998. Vertebral column morphology. C-start curvature, and the evolution of mechanical defenses intetraodontiform fishes. Copeia 1998:971–984.

Chapman GB, Dawson AB. 1961. Fine structure of larval anu-ran epidermis, with special reference to figures of Eberth.J Biophys Biochem Cytol 10:425–436.

Coughlin DJ, Valdes L, Rome LC. 1996. Muscle length changesduring swimming in scup: Sonomicrometry verifies the ana-tomical high-speed cine technique. J Exp Biol 199:459–463.

Danos N. 2005. Biomechanics of the skin during swimming inthe American Eel, Anguilla rostrata. Masters Thesis.Amherst: University of Massachusetts, 47 p.

D’Aout K, Curtin NA, Williams TL, Aerts P. 2001. Mechanicalproperties of red and white swimming muscles as a functionof the position along the body of the eel Anguilla anguilla.J Exp Biol 204:2221–2230.

Dingerkus G, Uhler DL. 1977. Enzyme clearing of Alcian blue-stained whole small vertebrates for demonstration of carti-lage. Stain Technol 32:229–231.

Donley JM, Dickson KA. 2000. Swimming kinematics of juve-nile Kawakawa Tuna (Euthynnus affinis) and Chub Mackerel(Scomber japonicus). J Exp Biol 203:3103–3116.

Donley JM, Shadwick RE. 2003. Steady swimming muscle dy-namics in the leopard shark Triakis semifasciata. J Exp Biol206:1117–1126.

Donley JM, Sepulveda CA, Konstantinidis P, Gemballa S, Shad-wick RE. 2004. Convergent evolution in mechanical design oflamnid sharks and tunas. Nature 429:61–65.

Donley JM, Shadwick RE, Sepulveda CA, Konstantinidis P,Gemballa S. 2005. Longitudinal patterns of red muscle strainand activation and body kinematics during steady swimmingin a lamnid shark, the shortfin mako (Isurus oxyrinchus).J Exp Biol 208:2377–2387.

Dowis JD, Sepulveda CA, Graham JB, Dickson KA. 2003.Swimming peformance studies on the eastern Pacific bonitoSarda chiliensis, a close relative of the tunas (family Scom-bridae). II. Kinematics. J Exp Biol 206:2749–2758.

Fishelson L. 1972. Histology and ultrastructure of the skin ofLepadichthys lineatus (Gobiesocidae: Teleostei). Mar Biol 17:357–364.

Fishelson L. 1996. Skin morphology and cytology in marine eelsadapted to different lifestyles. Anat Rec 246:15–29.

Fujii R. 1968. Fine structure of the collagenous lamella underly-ing the epidermis of the goby, Chasmichthys gulosus. AnnotZool Jpn 41:95–106.

Gemballa S. 2005. Locomotory specializations in aquatic envi-ronments: Morphological differences in the axial musculoten-dinous system of cruising specialists and accelerators. CompBiochem Physiol A 141:S163.

Gemballa S, Bartsch P. 2002. Architecture of the integument inlower teleostomes: Functional morphology and evolutionaryimplications. J Morphol 253:290–309.

Gemballa S, Hagen K. 2004. The myoseptal system in Chi-maera monstrosa and its evolution in the gnathostome stemlineage. Zoology 107:13–27.

Gemballa S, Roder K. 2004. From head to tail: The myoseptalsystem in basal actinopterygians. J Morphol 259:155–171.

Gemballa S, Treiber K. 2003. Cruising specialists and accelera-tors—Are different types of fish locomotion driven by differ-ently structured myosepta? Zoology 106:203–222.

Gemballa S, Vogel F. 2002. Spatial arrangement of white mus-cle fibers and myoseptal tendons in fishes. Comp BiochemPhysiol A 133:1013–1037.

Gemballa S, Ebmeyer L, Hagen K, Hoja K, Treiber K, Vogel F,Weitbrecht GW. 2003a. Evolutionary transformations of myo-septal tendons in gnathostomes. Proc R Soc Lond B270:1229–1235.

Gemballa S, Hagen K, Roder K, Rolf M, Treiber K. 2003b. Thehorizontal septum: Structure and evolutionary transforma-tions in vertebrates. J Evol Biol 16:966–975.

Gemballa S, Konstantinidis P, Donley JM, Sepulveda C, Shad-wick RE. 2006. Evolution of high-performance swimming insharks: transformation of the musculotendinous system from



subcarangidorm to thunniform swimmers. J Morphol 267:477–493.

Gemballa S, Kistner N, Sepulveda C, Bernal D. 2007. The mus-culotendinous system of pelagic fishes: How does swordfish(Xiphias gladius) compare to thunniform and carangifomswimmers? ICVM-8 abstract. J Morphol 268:1077.

Gillis GB. 1996. Undulatory locomotion in elongate aquatic ver-tebrates: Anguilliform swimming since Sir James Gray. AmZool 36:656–665.

Gillis GB. 1997. Anguilliform locomotion in an elongate sala-mander (Siren intermedia): Effects of speed on axial undula-tory movements. J Exp Biol 200:767–784.

Gillis GB. 1998a. Neuromuscular control of anguilliform locomo-tion: Patterns of red and white muscle activity during swim-ming in the American Eel Anguilla rostrata. J Exp Biol201:3245–3256.

Gillis GB. 1998b. Environmental effects on undulatory locomo-tion in the American Eel Anguilla rostrata: Kinematics onwater and land. J Exp Biol 201:949–961.

Gray J. 1933a. Studies in animal locomotion. I. The movementof fish with special reference to the eel. J Exp Biol 10:88–104.

Gray J. 1933b. Studies in animal locomotion. II. The relation-ship between waves of muscular contraction and the propul-sive mechanism of the eel. J Exp Biol 10:386–390.

Gray J. 1933c. Studies in animal locomotion. III. The propulsivemechanism of the whiting (Gadus merlangus). J Exp Biol10:391–400.

Gray J. 1968. Animal Locomotion. New York: WW Norton & Co.Harris JE, Crofton HD. 1957. Structure and function in the

nematodes—Internal pressure and cuticular structure inAscaris. J Exp Biol 34:116–130.

Hawkes JW. 1974. Structure of fish skin. I. General organiza-tion. Cell Tissue Res 149:147–158.

Hebrank MR. 1980. Mechanical properties and locomotor func-tions of eel skin. Biol Bull 158:58–68.

Hebrank MR. 1982. Functional morphology of fishes: Roles ofthe backbone and skin in locomotion. Doctoral Dissertation,Duke University. 142 p.

Hebrank MR, Hebrank JH. 1986. The mechanics of fish skin—Lack of an external tendon role in 2 teleosts. Biol Bull171:236–247.

Jayne BC. 1988. Mechanical-behavior of snake skin. J Zool214:125–140.

Katz SL, Shadwick RE, Rapoport HS. 1999. Muscle strain his-tories in swimming milkfish in steady as well as sprintinggaits. J Exp Biol 2002:529–541.

Katz SL, Syme DA, Shadwick RE. 2001. High speed swimming:Enhanced power in yellowfin tuna. Nature 410:770–771.

Kern S, Koumoutsakos P. 2006. Simulations of optimizedanguilliform swimming. J Exp Biol 209:4841–4857.

Lauder GV. 2006. Locomotion. In: Evans DH, Claiborne JB, edi-tors. The Physiology of Fishes, 3rd ed. Boca Raton: CRCPress. pp 3–46.

Lauder GV, Tytell ED. 2006. Hydrodynamics of undulatory pro-pulsion. In: Shadwick RE, Lauder GV, editors. Fish Biome-chanics. Fish Physiology, Vol. 23. San Diego: Academic Press.pp 425–468.

Le Guellec D, Morvan-Dubois G, Sire JY. 2004. Skin develop-ment in bony fish with particular emphasis on collagen depo-sition in the dermis of the zebrafish (Danio rerio). Int J DevBiol 48:217–231.

Leonard JB, Summers RG. 1976. Ultrastructure of integumentof American eel. Anguilla rostrata. Cell Tissue Res 171:1–30.

Liem KF. 1967. Functional morphology of integumentary respi-ratory and digestive systems of synbranchoid fish Monopterusalbus. Copeia 1967:375–388.

Lindsey CC. 1978. Form, function, and locomotory habits infish. In: Hoar WS, Randall DJ, editors. Locomotion. FishPhysiology, Vol. 7. New York: Academic Press. pp 1–100.

Logan VH, Odense PH. 1974. Integument of ocean sunfish (Mola-Mola L) (Plectognathi) with observations on lesions from twoectoparasites. Capsala-Martinierei (Trematoda) and Philortha-goriscus-Serratus (Copepoda). Can J Zool 52:1039–1045.

Long JH Jr. 1998. Muscles, elastic energy, and the dynamics ofbody stiffness in swimming eels. Am Zool 38:771–792.

Long JH Jr, Hale ME, McHenry MJ, Westneat MWW. 1996.Functions of fish skin: Flexural stiffness and steady swim-ming of longnose gar Lepisosteus osseus. J Exp Biol 199:2139–2151.

Long JH Jr, Adcock B, Root RG. 2002. Force transmission viaaxial tendons in undulating fish: A dynamic analysis. CompBiochem Physiol A 133:911–929.

Mittal AK, Banerjee TK. 1974. Structure and keratinization ofthe fresh water teleost Notopterus notopterus (Notopteridae,Pisces). J Zool (Lond) 174:341–355.

Mittal AK, Banerjee TK. 1975a. Histochemistry and structureof skin of a Murrel, Channa striata (Bloch, 1797) (Channi-formes, Channidae). I. Epidermis. Can J Zool 53:833–843.

Mittal AK, Banerjee TK. 1975b. Histochemistry and structureof skin of a Murrel, Channa striata (Bloch, 1797) (Channi-formes, Channidae). II. Dermis and subcutis. Can J Zool53:844–852.

Motta PJ. 1977. Anatomy and functional-morphology of dermalcollagen-fibers in sharks. Copeia 1977:454–464.

Muller UK, Smit J, Stamhuis EJ, Videler JJ. 2001. How thebody contributes to the wake in undulatory swimming fish:Flow fields of a swimming eel (Anguilla anguilla). J Exp Biol204:2751–2762.

Olsson R. 1961. The skin of Amphioxus. Z Zellforsch 54:90–104.Patterson C, Johnson GD. 1995. The intermuscular bones and

ligaments of teleostean bones. Smithsonian Contrib Zool559:1–83.

Romeis B. 1986. Mikroskopische Technik. Munchen: Urban andSchwarzenberg. 697 p.

Shadwick RE, Gemballa S. 2006. Kinematics and muscle dy-namics in undulatory swimming. In: Shadwick RE, LauderGV, editors. Fish Biomechanics. Fish Physiology, Vol. 23. SanDiego: Academic Press. pp 241–280.

Shadwick RE, Steffensen JF, Katz SL, Knower T. 1998. Muscledynamics in fish during steady swimming. Am Zool 38:755–770.

Shadwick RE, Katz SL, Korsmeyer K, Knower T, Covell JW.1999. Muscle dynamics in skipjack tuna Katsuwonus pela-mus: Timing of red muscle shortening and body curvatureduring steady swimming. J Exp Biol 202:2139–2150.

Tytell ED. 2004a. The hydrodynamics of eel swimming. II.Effect of swimming speed. J Exp Biol 207:3265–3279.

Tytell ED. 2004b. Kinematics and hydrodynamics of linearacceleration in eels, Anguilla rostrata. Proc R Soc Lond B271:2535–2540.

Tytell ED, Lauder GV. 2004. The hydrodynamics of eel swim-ming. I. Wake structure. J Exp Biol 207:1825–1841.

Videler JJ. 1993. Fish Swimming. Fish and Fisheries Series10. London: Chapman & Hall. 260 p.

Vogel S. 2004. Comparative Biomechanics. Princeton andOxford: Princeton University Press.

Wardle CS, Videler JJ, Altringham JD. 1995. Tuning in to fishswimming waves: Body form, swimming mode and musclefunction. J Exp Biol 198:1629–1636.

Westneat MW, Wainwright SA. 2001. Mechanical design forswimming: Muscle, tendon, and bone. In: Block BA, StevensED, editors. Tuna: Physiology, Ecology, and Evolution. FishPhysiology, Vol. 19. San Diego: Academic Press. pp 271–311.

Westneat MW, Hale ME, McHenry MJ, Long JH Jr. 1998.Mechanics of the fast-start: Muscle function and the role ofintramuscular pressure in the escape behavior of Amia calvaand Polypterus palmas. J Exp Biol 201:3041–3055.

Wainwright SA. 1983. To bend a fish. In: Webb PW, Weihs D,editors. Fish Biomechanics. New York: Praeger Publishers. pp68–91.

Wainwright SA, Vosburgh F, Hebrank JH. 1978. Shark skin—Function in locomotion. Science 202:747–749.

Wainwright SA, Biggs WD, Currey JD, Gosline JM. 1982. Me-chanical Design in Organisms. Princeton, New Jersey: Prince-ton University Press.

44 N. DANOS ET AL.


The Musculotendinous System of an Anguilliform Swimmer ...glauder//reprints_unzipped/Danos et...

Documents

Transcript of The Musculotendinous System of an Anguilliform Swimmer ...glauder//reprints_unzipped/Danos et...