Comparative Biochemistry and Physiology Part A 133(2002) 1159–1170

1095-6433/02/$ - see front matter� 2002 Elsevier Science Inc. All rights reserved.PII: S1095-6433Ž02.00241-6

Review

The evolution of tendon — morphology and material properties�

Adam P. Summers *, Thomas J. Kooba, b

Ecology and Evolutionary Biology, 321 Steinhaus Hall, University of California, Irvine, CA 92697-2525, USAa

Shriners Hospital for Children, 12502 North Pine Drive, Tampa, FL 33612, USAb

Received 25 February 2002; received in revised form 19 August 2002; accepted 19 August 2002

Abstract

Phylogenetically, tendinous tissue first appears in the invertebrate chordateBranchiostoma as myosepta. This two-dimensional array of collagen fibers is highly organized, with fibers running along two primary axes. In hagfish the firstlinear tendons appear and the myosepta have developed specialized regions with unidirectional fiber orientation—a lineartendon within the flat sheet of myoseptum. Tendons react to compressive load by first forming a fibrocartilaginous pad,and under severe stress, sesamoid bones. Evidence for this ability to react to load first arises in the cartilaginous fish,here documented in a tendon from the jaw of a hard-prey crushing stingray. Sesamoid bones are common in bony fishand also in tetrapods. Tendons will also calcify under tensile loads in some groups of birds, and this reaction to load isseen in no other vertebrates. We conclude that the evolutionary history of tendon gives us insight into the use of modelsystems for investigating tendon biology. Using mammal and fish models may be more appropriate than avian modelsbecause of the apparent evolution of a novel reaction to tensile loads in birds.� 2002 Elsevier Science Inc. All rights reserved.

Keywords: Stress; Strain; Tensile; Collagen; Hagfish; Agnathans; Strength; Stiffness; Morphology; Material properties; Model systems;Connective tissue; Evolutionary transitions; Vertebrata

‘‘Tendons, forming the attachment of many muscles,consist of bundles of connective tissue fibers’’

Alfred S. Romer.

1. Introduction

Tendon is most often thought of as the brightwhite, parallel fibered connective tissue that joins

� This paper was presented at ‘Tendon – Bridging the Gap’,a symposium at the 2002 Society of Integrative and Compar-ative Biology. Participation was funded by SICB, The ShrinersHospitals for Children, and the National Science Foundation(IBN-0127260).*Corresponding. Tel.:q1-949-824-9359; fax:q1-510-315-

3106.E-mail addresses: asummers@uci.edu(A.P. Summers),

tkoob@shctampa.usf.edu(T.J. Koob).

muscle to bone, but this definition does not beginto account for the varied forms of tendinoustissue—even within a single organism—its adap-tation to load, or the form that it takes in lowervertebrates. In order to understand the evolutionof the musculoskeletal system in vertebrates it isimportant to understand the evolutionary historyof tendon as a tissue. Our understanding of tendonhas lagged behind that of bone and cartilage, bothof which have been placed in an evolutionarycontext (Moss and Moss-Salentijn, 1983; Person,1983; Smith and Hall, 1990). In fact, comparativeresearch on tendon is so rare that this project maybe premature. Nevertheless, we attempt in the nextfew pages to synthesize tendon research in anevolutionary context, if for no other purpose than

1160 A.P. Summers, T.J. Koob / Comparative Biochemistry and Physiology Part A 133 (2002) 1159–1170

to point out those areas that sorely need furtherinvestigation.We consider tendon to be a tissue, and will refer

to it as such, though at times we will use ‘tendi-nous tissue’ to make clear the distinction betweenthe tissue and a particular tendinous organ(i.e.digital flexor tendon). Our working definition oftendon is an amalgam of functional and composi-tional definitions: a dense, highly organized tissuethat transmits force, and is composed predomi-nantly of type I collagen with fibril associatedproteoglycans.

2. Morphology

Tendon is most familiar as the densely packed,linearly-arrayed, collagenous tissue seen in flexorand extensor tendons in tetrapod limbs. However,it should be recognized that there are other collag-enous connective tissues that perform the samefunctional task, that of transferring force generatedby muscle from the muscle end to a target. Theseinclude the dense, well-organized arrays of con-nective tissue in the myosepta of lower vertebrates,as well as aponeuroses.The body musculature of vertebrates is arranged

in serially homologous myomeres, each separatedfrom the next by a collagenous myoseptum. Inhumans one of the few vestiges of this segmenta-tion is the ‘six-pack’ of the abdominal muscles,but as anyone who has eaten a fish fillet can attest,blocky myomeres are far more obvious in lowervertebrates. Myosepta are not mats of disorganizedcollagen but rather highly organized structures withcollagen organized in uni- or bi-directional sheets(Liem et al., 2000; Vogel and Gemballa, 2000).Furthermore, it should be recognized that simple

myosepta and linearly-arrayed tendon are extremesin a continuum. There are numerous examplesfrom fish and tetrapods of aponeuroses that gradeinto tendon(Liem et al., 2000), and, within themyosepta of fish there are collagen fibers withpreferred orientation that form linear tendon struc-tures(Westneat et al., 1993; Vogel and Gemballa,2000; Gemballa and Vogel, 2002).In this section we will examine morphological

innovations in these dense, collagenous tissues thatserve to transmit the force of muscles across anarray of chordate taxa. In the interest of brevitywe will establish a plesiomorphic condition fortendinous tissue and then specifically treat only

those taxa that exhibit novel morphology relativeto this ‘ancestral’ state.

2.1. Urochordates

The most well known group of urochordates,the ascidians or sea-squirts, are sessile as adults,but as larvae they may be free-swimming ‘tad-poles’ propelling themselves along a helical pathwith undulations of their tail(Gans et al., 1996;McHenry, 2001). The muscles that drive this loco-motion are not divided by collagenous myosepta,and the muscle itself more closely resembles car-diac than striated muscle(Cloney and Burighel,1991). The only hint of tendinous tissue is in aseries of collagenous, connective tissue leafletsthat issue from the notochordal sheath but do notappear to be attached to muscle(Cloney, 1964).At present there is no evidence for tendinousstructures in urochordates.

2.2. Cephalochordates

The undulatory locomotion of the lancelets ispowered by myotomal muscle divided by collag-enous myosepta(Fig. 1) (Weitbrecht, 2000), afeature that is clearly seen in several fossil formsincludingPikaia from the Burgess Shale.There is scant evidence of any linear tendons in

lancelets, the only possible exception being struc-tures that Ruppert(1991) refers to as ‘micro-ligaments’. These are connective tissue projectionsthat join the myomeric muscle to the body wall,however, they are quite thin(1–5 mm) and thereis no evidence that they are collagenous. We inferthat well-organized and functional fibrous myosep-ta appear for the first time in the chordate lineagein Cephalochordates.

2.3. Agnathans

The jawless fish, lampreys and hagfish usemyomeric muscle to power undulatory locomotion,and according to the well preserved fossils ofHaikouichthys andMyllokunmingia, this has beenthe case since the Lower Cambrian(490–540MYA ) (Chen et al., 1999; Shu et al., 1999). Themyosepta are collagenous sheets with evidence forthe local specializations that form linearly arrayedtendons in bony fish(Vogel and Gemballa, 2001).Hagfish also have an interesting hydrostatic tongue

1161A.P. Summers, T.J. Koob / Comparative Biochemistry and Physiology Part A 133 (2002) 1159–1170

Fig. 1. A light micrograph of a single myseptum fromBran-chiostoma lanceolatum shows the organization of the collagenfibers into a two-dimensional array. The inscribed lines are toindicate the two primary directions of the collagen fibers(mod-ified with permission from Weitbrecht, 2000).

protrusion and retraction mechanism that involvestwo pennate muscles and two linear tendons(Fig.2) (Coles, 1905, 1907, 1914). We have preparedhistological sections(10 mm paraffin) of one ofthe tongue retractor tendons(M. retractor mandi-buli tendon). The tendon is superficially quitesimilar in appearance to that of mammals, havingthe characteristic bright white color, clearly orient-ed fibrous material, and discrete edges. However,it appears to lack the organization into primaryand secondary bundles, as well as the epitenonand paratenon seen in mammals(Kastelic et al.,1978). The only indication of this hierarchicalorganization is that there is an inner ‘core’ tendonwrapped with an outer ‘sheath’ tendon(Fig. 3).

2.4. Cartilaginous and bony fish

Cartilaginous and bony fish have tendinousmyosepta between the blocks of body musculatureas well as linear tendons associated with jaw andpharyngeal muscles. The former appear quite sim-ilar to the myosepta in agnathans with somespecialized regions that are linear in orientation(Hagen et al., 2001). The linear tendons found inthe feeding apparatus resemble those of mammalsand some individual tendons are likely homologsof mammalian tendons(i.e. an adductor mandi-bularis tendonstemporalis tendon) (Liem et al.,2000). Furthermore, the cartilaginous fish showtendon specializations that have previously onlybeen reported in tetrapods. Tendons are normallysubjected to tensile loads, however when onemakes a sharp turn around a corner it is alsosubjected to compressive and shear loading in theregion of the bend. This compressive loading leadsto the formation of a fibrocartilaginous pad alongthe inner surface of the bend(Vogel and Koob,1989; Benjamin and Ralphs, 1998).Some stingrays eat hard prey almost exclusively,

and have several specializations of their jaws thatreinforce their mandibular cartilage and increasethe effective force of their jaw adductors(Sum-mers et al., 1998; Summers, 2000). The ventralintermandibular tendon of one of these specialists,the cownose ray(Rhinoptera bonasus), makes asharp turn as it passes dorsally around the cornerof the jaws. We found a fibrocartilaginous padassociated with the tendon along the inner edge ofthis sharp bend(Fig. 4a). Histologically it appearssimilar to fibrocartilage from mammals(Fig. 4b)and we believe this is the phylogenetically earliestexample of tendon’s ability to adapt to compres-sive load. The ability of tendon to respond tocompressive load by forming a fibrocartilaginouspad has been confirmed in amphibians(Carvalhoand Vidal, 1994, 1995), archosaurs(Felisbino andCarvalho, 1999), and mammals(for review seeBenjamin and Ralphs, 1998).A further adaptation to compressive load is the

formation of sesamoid bones in regions of tendonthat bend over a bony prominence(for review seeSarin et al., 1999). These bones arise within thetendon and may be encapsulated in a synovialsheath. In mammals there is evidence that sesa-moid formation can be induced by load, howeversome sesamoids form regardless of force regime

1162 A.P. Summers, T.J. Koob / Comparative Biochemistry and Physiology Part A 133 (2002) 1159–1170

Fig. 2. Panels 1–4 show the retraction of the toothy tongue ofMyxine glutinosa accomplished by manually pulling on the retractortendon. Scale bars5 mm. The lower panel shows the anatomy of the teeth, the supporting cartilage and the two tendons that protractand retract the tooth plate(modified from Coles, 1907).

(i.e. the patella). There are no sesamoids in carti-laginous fish, but they are prominent in bony fish.The urohyal, the bone that supports the tongue,may be the most well known example, thoughother bones of the branchial arches are also formedin tendon and are clear in fossil and recent taxa(Patterson, 1977; Arratia and Schultze, 1990; Liemet al., 2000). Sesamoids are widespread in othervertebrates including amphibians(Olson, 2000;Trueb et al., 2000), squamate reptiles(Maisano,2001), archosaurs (Sych, 2001; Hutchinson,2002), and mammals.

2.5. Archosaurs

In addition to the aforementioned response to acompressive load there are indications that tendonwill respond to tensile loads by mineralizing(forreview see Landis, 2002). These mineralizationsare quite different from sesamoid bone, in thatthey never become encapsulated in a synovialenvelope, they do not have the histology of bone,and they are not associated with tendons passingover bony prominences. Rather these mineraliza-tions usually arise when a tendon bifurcates, lead-

1163A.P. Summers, T.J. Koob / Comparative Biochemistry and Physiology Part A 133 (2002) 1159–1170

Fig. 3. (a) An alcian blue stained longitudinal section throughthe tongue retractor tendon approximately midway along itslength. The gaps in the dense connective tissue separate theinner ‘core’ tendon from the outer ‘sheath’ tendon and are theonly indication of tendon organization. Scale bars200mm. (b)An hemotoxylin and eosin stained section showing the rela-tively high cellularity of the gap region compared to the ten-dinous region. There are few tenocytes evident, as is often thecase in mammalian tendon. Scale bars125mm.

ing to a higher load in the two smaller tendonsthan there is in the conjoined tendons. They appearto have arisen in the non-avian dinosaurs(Hutch-inson, 2002), but as far as we can tell are foundin no other group of chordates.

2.6. Mammals

Aside from the hierarchical organization ofmammalian tendon(Kastelic et al., 1978), thereare no morphological characters of tendon that wedo not see in other vertebrate groups. A possibleexception may be in the morphology of the tendonentheses, the insertion of the tendon onto the bone,but there are no comparative data from other taxa(Benjamin et al., 2002).

3. Material properties

There are two principal difficulties in synthesiz-ing an evolutionary perspective on the materialproperties of tendon: firstly, there is no acceptedstandard protocol for testing, so results of differentstudies are difficult to compare; second, there arefew literature values for the mechanical propertiesof tendon from lower vertebrates, and none at allfrom cartilaginous fish or agnathans. To addressthe first issue we present a range of representativepublished values for several biomechanicallyimportant measures of the response of tendon toapplied force. To provide a broader evolutionarypicture we also present original data from our owntests on a relatively obscure, linearly arrayed ten-don from the hagfish described above.Biological structures are notoriously difficult to

characterize with the standard material properties,such as strength and stiffness, used for manufac-tured solids. The reasons for this fall into twocategories—technical problems that have theirbasis in the inconvenient sizes and shapes of livingtissue; and difficulties that arise from the hetero-geneous make-up and visco-elasticity of naturalmaterials(Vogel, 1988; Niklas, 1992; Martin etal., 1998). Tendons present special technical chal-lenges from the point of view of clamping andpreconditioning, and are certainly heterogeneous,viscoelastic materials, nevertheless there is anextensive literature on the tensile properties ofmammalian tendon(i.e. Bennett et al., 1986;Pollock and Shadwick, 1994b; de Zee et al., 2000).

3.1. Hagfish tendon

We dissected 14, fresh, frozen hagfish(Myxineglutinosa) heads, removing the completeretractormandibuli tendon from its origin on a thin carti-laginous rod at the posterior end of the tongueretractor muscle to the insertion on the cartilagi-nous tooth plate(for relevant anatomy see Coles,1907). Tendons were kept moist with hagfishRingers(0.53 M NaCl, 7 mM KCl, 4.5 mM CaCl,12 mM MgCl , 0.9 mM Na SO , 25 mM Tris at2 2 4

pH 7.95) (Riegel, 1978) for the brief time betweendissection and testing. Material properties weredetermined with a tensile test to failure using anMTS Mini Bionix with a 500-g load cell at astrain rate of 2 mmys. Approximately 1 cm ofeach end of the tendon was held between the

1164 A.P. Summers, T.J. Koob / Comparative Biochemistry and Physiology Part A 133 (2002) 1159–1170

Fig. 4.(a) A ventral view of the upper and lower jaws of a cownose ray(Rhinoptera bonasus) showing the arrangement of the adductormandibularis medialis(AMM ) muscle and the associated tendon.(b) Masson’s trichrome stained paraffin section(10 mm thick) ofthe fibrocartilaginous pad from the AMM tendon showing regions of proteoglycan matrix(blue) and organized fibrous areas(red).Scale bars100mm. (c) Section of the fibrocartilaginous region of a calf meniscus showing similar morphology to the stingray tendon.Scale bars100mm

serrate jaws of standard tensile grips. Grips werethen cooled by manual application of dry ice(–78 8C) until the tissue inside the grips wasvisibly affected. Tissue between the grips(8–15mm) was checked with forceps to ensure that ithad not frozen. In spite of freeze clamping, thetendon shifted in the grips in 6 of 14 tests resultingin slippage artifacts that prevented us from usingthese samples in calculating the maximum strainof the specimen(Fig. 5). The diameter of theunstretched, cylindrical tendon was measured with

calipers to the nearest tenth of a millimeter. Stressvs. strain curves were generated from the MTSdata and used to calculate ultimate strength, stiff-ness, and failure strain.The strength of the hagfish tongue retractor

tendon was 47.8"3.5 MPa, and from the samplesthat did not slip in the clamps(ns8) we calculatedthe strain at failure to be 22"2%. This value iswell beyond any reliable estimate of maximumstrain in mammalian tendon(-10%). The stiff-ness of the tendon was 290"29 MPa. These values

1165A.P. Summers, T.J. Koob / Comparative Biochemistry and Physiology Part A 133 (2002) 1159–1170

Fig. 5. Stress vs. strain curves for the hagfish(Myxine glutinosa) tongue retractor tendon. The ranges of values for strength and breakingstrain in mammalian tendon are indicated by the bars.

are similar to values reported for some mammaliantendons(Fig. 5).

3.2. Bony and cartilaginous fish

Cartilaginous fish have few linearly arrayedtendons(Liem and Summers, 1999), and there hasbeen no assessment of their material properties,nor the properties of the aponeurotic tendons. Theliterature for bony fish contains just two datapoints. In this volume Shadwick et al.(2002)report that the tail tendons of albacore and yellow-fin tuna, fast swimming scombrid fish, are quitestiff (1.19–1.43 GPa), and they estimate a break-ing strength near 30 MPa.

3.3. Amphibians

Of the three extant groups of lissamphibians(frogs, salamanders, and caecilians) only frog ten-dons have been tested in depth. There are no

records for the limbless and enigmatic caecilians,and a single record for salamanders—Azizi et al.(2002) found a maximum stiffness of 7.1 MPa forthe sheet-like tendinous myosepta of the siren(Siren lacertina). Although frogs are importantmodel systems for the study of muscle only a fewstudies address tendon mechanics, and those onlyreport values for stiffness(Lieber et al., 1991;Trestik and Lieber, 1993). Gastrocnemius tendon(Es1.5 GPa) is similar to mammals, but thesemitendinosus(0.2 GPa) is only as stiff as hagfishtendon.

3.4. Archosaurs

Of the two extant groups of archosaurs, the birdsand the crocodilians, there are data only from theformer, though the latter have prominent, function-ally important tendons. Some groups of birds(andsome extinct non-avian dinosaurs; Hutchinson,2002) mineralize their tendons, which has a pro-

1166 A.P. Summers, T.J. Koob / Comparative Biochemistry and Physiology Part A 133 (2002) 1159–1170

Fig. 6. Scattergram showing values for(a) stiffness and(b)strength of tendon from several chordate taxa. The values forhagfish(Myxine glutinosa) are from this paper. Data for fish(Shadwick et al., 2002), amphibians(Lieber et al., 1991, 2000;Trestik and Lieber, 1993), birds (Bennett and Stafford, 1988;Leonard et al., 1976), and mammals(Bennett et al., 1986; Itoiet al., 1995; Kondo et al., 1998; Pollock and Shadwick, 1994b;Wren et al., 2001) were gathered from literature sources.

Fig. 7. Cladogram of the chordates(modified from Liem et al.,2000) showing the evolutionary transitions of tendon:(1)‘leaflets’ of connective tissue emanating from the notochordsheath;(2) an organized array of collagen forms myosepta;(3)linear tendon; specializations in the myosepta;(4) tendonresponds to compressive loading with a fibrocartilaginous pad;and(5) sesamoid bones in areas of high compressive loading.

found effect on their properties(Bennett and Staf-ford, 1988). Indeed in some cases the stiffness ofcalcified tendon can approach that of bone(Leon-ard et al., 1976; Bennett and Stafford, 1988). Herewe are interested in the unmineralized tendon’sbehavior under load, which has been measured inthe usual tensometer devices as well as in vivo,and in vitro using disparate techniques such assonomicrometry, force buckles, and ultrasound(Biewener et al., 1992; Mohamed et al., 1995;Buchanan and Marsh, 2001). The stiffness of aviantendons(mineralized and unmineralized) rangedfrom 0.9 to 13 GPa, and the strength from 54 to117 MPa.

3.5. Mammals

There is a vast body of literature on the tech-niques of testing mammalian tendon(i.e. Riemers-ma and Schamhardt, 1982; Martin et al., 1998;Ker et al., 2000), and on the properties of tendon(i.e. Bennett et al., 1986; Pollock and Shadwick,

1994b; Pike et al., 2000). From a functional pointof view mammalian tendons fall into two catego-ries, those that act as springs, such as the Achillesand digital flexor tendons of quadrupedal mam-mals, and those that simply transmit muscle forcewithout storing significant energy, as in the digitalextensors and biceps tendon(Alexander, 1984;Biewener and Blickhan, 1988; Biewener, 1998).Oddly, though these two functions might best beaddressed by varied material properties, it seemsthat natural selection has not led to significantdifferences between them(Pollock and Shadwick,1994b). In other words, while it would seemlogical that the amount of strain energy stored ina tendon could be ‘tuned’ by varying the stiffnessand resilience, in fact these properties are relativelyinvariant across a wide variety of tendons.Rather than rehash these studies, or a subset of

them, we would like to point out the high and lowvalues that have been reported for several proper-ties and present scattergrams that summarize asynoptic review of the literature(Fig. 6).

4. Discussion

Though there is too little evidence from non-mammalian taxa to be completely confident in anevolutionary scenario, we would propose the fol-lowing hypotheses(Figs. 7 and 8). The firstappearance of tendinous tissue is as myosepta, thin

1167A.P. Summers, T.J. Koob / Comparative Biochemistry and Physiology Part A 133 (2002) 1159–1170

Fig. 8. Cladogram of the vertebrates(modified from Liem etal., 2000) showing the evolutionary transition of tendon:(1)tendon responds to compressive loading with a fibrocartilagi-nous pad;(2) sesamoid bones in areas of high compressiveloading; and(3) tendons in tension calcify in response to load.

sheets of connective tissue with collagen fibersorganized to resist tension from several directions.As body plans became more adapted for high-speed undulatory locomotion the myosepta becamespecialized, with some regions being composedalmost entirely of unidirectional collagen fibers—a linear tendon within the flat sheet of the myosep-tal tendon. This is supported by the work ofGemballa’s group onBranchiostoma, agnathans,and fish (Vogel and Gemballa, 2001; Weitbrechtand Gemballa, 2001; Gemballa and Vogel, 2002).Linear tendons appear in the agnathans(Acrania)before the advent of vertebrates, and before theevolution of endochondral bone(Smith and Hall,1990). This renders the textbook definition oftendon as a tissue that connects muscle to bonesomewhat problematic.Sometime in or before the Silurian(440 to 410

mya) tendon acquired the ability to adapt tocompressive loads by forming fibrocartilage, andby the Devonian, bony fish have formed sesamoidbones (Patterson, 1977; Arratia and Schultze,1990). This is supported by our own work on thefibrocartilage of a cartilaginous fish(see precedingsection) and by the fossil record of fish(Long,1995). The ability to calcify tendon in response totensile loads is not certainly present until theevolution of birds in the Jurassic period(208 to146 mya) (Padian, 1997). The uniformity in mate-rial properties and the lack of novel responses toloading in eutherian and metatherian mammals is

an indication that tendon has not undergone anymajor evolutionary transitions in the mammals.This implies that some non-mammals, and evennon-tetrapods, may be used as model systems forstudying tendon development, physiology, mor-phology, and repair. It is important to note thatbirds may not be a good model system for mam-malian tendon biology, as avian tendon appears tohave evolved a novel response to tensile stress(Fig. 8).On the evolution of material properties there is

little we can be sure of, but we would speculatethat the high strength of linear tendon is primitiveon the basis of our hagfish data. Stiffness mayhave increased over evolutionary time in the line-ages that gave rise to fish and tetrapods, butstrength arose early and has been little changedthrough evolutionary time. We require a moreextensive survey of material properties in fish andlower tetrapods to generate further hypotheses.Clearly, we have woefully little knowledge of

non-mammalian tendon. Among the significantgroups for which there are little or no data are thesquamate reptiles, turtles, and cartilaginous fish.This is particularly troubling given the interestingtransition that apparently takes place between theagnathans and the tetrapods—tendon becomesstiffer, it gains an epitenon, and becomes integratedinto the locomotor system as a spring. We cannotunderstand the current character states in mammalswithout learning the evolutionary trajectory thatthey traversed in getting there.In the area of material properties there are

several topics that would be quite interesting toknow more about. The reported values for resil-ience, that is the efficiency of energy storage, formammalian and fish tendon are quite high—onthe order of 90%(Alexander, 1984; Pollock andShadwick, 1994a; Shadwick et al., 2002). The fishtendons and mammalian tendons that have beentested are of similar stiffness(Fig. 6). From thescant comparative data it is clear that there aretendons of far lesser stiffness in both frogs andhagfish. It would be interesting to know if thehigh resilience is linked to high stiffness or wheth-er these traits are independent.The lack of an epitenon in the hagfish tongue

tendon may represent a primitive, less hierarchicalstructure. Comparative morphological data frombony and cartilaginous fish, amphibians and rep-tiles would settle the question of where the mam-malian architecture of tendon arose(Kastelic et

1168 A.P. Summers, T.J. Koob / Comparative Biochemistry and Physiology Part A 133 (2002) 1159–1170

al., 1978; Rowe, 1985), and concomitantly mightshow whether the increased stiffness of mamma-lian tendon relative to hagfish tendon is related tothis organization.Perhaps the most exciting area in tendon

mechanics is that of the viscoelastic properties andfatigue quality(Ker et al., 2000; Ker, 2002). Withonly a few data from sheep and kangaroos, thereremains a gap in our understanding of the evolu-tion of time-dependent responses to steady andcyclic loading. This lack of knowledge is particu-larly troubling given the wide range of locomotorbehaviors and novel stress regimes that appear in´the tetrapod lineage. The evolutionary history offatigue quality and creep responses may be linkedto the evolution of terrestriality, a cursorial life-style, or even metabolic rate.

Acknowledgments

Magdalena Koob-Emunds performed histologi-cal chores on short notice with good spirit. EricHilton provided fossil advice, and the paper ben-efited from discussions with Beth Brainerd, andmany other participants in the SICB symposium‘Tendon-Bridging the Gap’. We particularly thankSven Gemballa for the lancelet myosepta photo-graph. Steve Kajiura kindly commented on anearlier draft of this manuscript. The McDowellFoundation, The Shriners Hospitals for Children,and the National Science Foundation(IBN-0127260) provided support for this work.

References

Alexander, R.M., 1984. Elastic Mechanisms in Animal Move-ment. Cambridge University Press, Cambridge.

Arratia, G., Schultze, H.-P., 1990. The urohyal: developmentand homology within osteichthyans. J. Morphol. 203,247–282.

Azizi, E., Gillis, G.B., Brainerd, E.L., 2002. Morphology andmechanics of myosepta in a swimming salamander(Sirenlacertina). Comp. Biochem. Physiol. 133A, 967–978.

Benjamin, M., Kumai, T., Milz, S., Boszczyk, B.M., Boszczyk,A.A., Ralphs, J.R., 2002. The skeletal attachment of ten-dons—tendon ‘entheses’. Comp. Biochem. Physiol. 133A,931–945.

Benjamin, M., Ralphs, J.R., 1998. Fibrocartilage in tendonsand ligaments—an adaptation to compressive load. J. Anat.193, 481–494.

Bennett, M.B., Ker, R.F., Dimery, N.J., Alexander, R.M., 1986.Mechanical properties of various mammalian tendons. J.Zool. Lond. 209, 537–548.

Bennett, M.B., Stafford, J.A., 1988. Tensile properties ofcalcified and uncalcified avian tendons. J. Zool. Lond. 214,343–351.

Biewener, A.A., 1998. Muscle–tendon stresses and elasticenergy storage during locomotion in the horse. Comp.Biochem. Physiol. 120A, 73–87.

Biewener, A.A., Blickhan, R., 1988. Kangaroo rat locomotion:design for elastic energy storage or acceleration? J. Exp.Biol. 140, 243–255.

Biewener, A.A., Dial, K.P., Goslow, G.E., 1992. Pectoralismuscle force and power output during flight in the starling.J. Exp. Biol. 164, 1–18.

Buchanan, C.I., Marsh, R.L., 2001. Effects of long-termexercise on the biomechanical properties of the Achillestendon of guinea fowl. J. Appl. Physiol. 90, 164–171.

Carvalho, H.F.D., Vidal, B.D.C., 1994. Structure and histo-chemistry of a pressure-bearing tendon of the frog. Ann.Anat. 176, 161–170.

Carvalho, H.F.D., Vidal, B.D.C., 1995. The elastic system ofa pressure-bearing tendon of the bullfrogRana catesbeiana.Ann. Anat. 177, 397–404.

Chen, J.Y., Huang, D.Y., Li, C.W., 1999. An early Cambriancraniate-like chordate. Nature(Lond.) 402, 518–522.

Cloney, R.A., 1964. Development of the ascidian notochord.Acta Embryol. Morphol. Exp. 7, 111–130.

Cloney, R.A., Burighel, P., 1991. Ascidiacea. In: Harrison,F.W., Ruppert, E.E.(Eds.), Microscopic Anatomy of Inver-tebrates. Wiley-Liss, New York, pp. 221–345.

Coles, F.J., 1905. A monograph on the general morphology ofthe myxinoid fishes, based on a study of Myxine. Part I.The anatomy of the skeleton. Trans. R. Soc. Edinburgh 41,749–788.

Coles, F.J., 1907. A monograph on the general morphology ofthe myxinoid fishes, based on a study of Myxine. Part II.The anatomy of the muscles. Trans. R. Soc. Edinburgh 45,683–757.

Coles, F.J., 1914. A monograph on the general morphology ofthe myxinoid fishes, based on a study of Myxine. Part III.Further observations on the skeleton. Trans. R. Soc. Edin-burgh 46, 669–681.

de Zee, M., Bøjsen-Moller, F., Voigt, M., 2000. Dynamic¨viscoelastic behavior of lower extremity tendons duringsimulated running. J. Appl. Physiol. 89, 1352–1359.

Felisbino, S.L., Carvalho, H.F., 1999. Identification and distri-bution of type VI collagen in tendon fibrocartilages. J.Submicrosc. Cytol. Pathol. 31, 187–195.

Gans, C., Kemp, N.E., Poss, S.G., 1996. Lancelets(Cephalo-chordata): A New Look at Some Old Beasts. Laser PagesPub, Jerusalem, Israel.

Gemballa, S., Vogel, F., 2002. The spatial arrangement ofwhite muscle fibers and myoseptal tendons in fishes. Comp.Biochem. Physiol. 133A, 1013–1037.

Hagen, K., Gemballa, S., Freiwald, A., 2001. Locomotorydesign and locomotory habits ofChimaera monstrosa. J.Morphol. 248, 237.

Hutchinson, J.R., 2002. The evolution of hindlimb tendonsand muscles on the line to crown-group birds. Comp.Biochem. Physiol. 133A, 1051–1086.

Itoi, E., Berglund, L.J., Grabowski, J.J., Schultz, F.M., Grow-ney, E.S., Morrey, B.F., An, K.-N., 1995. Tensile propertiesof the supraspinatus tendon. J. Orthop. Res. 13, 578–584.

Kastelic, J., Galeski, A., Baer, E., 1978. The multicompositestructure of tendon. Conn. Tissue Res. 6, 11–23.

1169A.P. Summers, T.J. Koob / Comparative Biochemistry and Physiology Part A 133 (2002) 1159–1170

Ker, R.F., 2002. The implications of the adaptable fatiguequality of tendond for their construction, repair and function.Comp. Biochem. Physiol. 133A, 987–1000.

Ker, R.F., Wang, X.T., Pike, A.V.L., 2000. Fatigue quality ofmammalian tendons. J. Exp. Biol. 203, 1317–1327.

Kondo, E., Yasuda, K., Miyata, K., Hara, N., Kaneda, K.,1998. The mechanical properties of the semitendinosus andgracilis tendons. Hokkaido J. Orthop. Traum. 40, 13–15.

Landis, W., Silver, F.H., 2002. The structure and function ofnormally mineralizing avian tendons. Comp. Biochem. Phy-siol. 133A, 1135–1157.

Leonard, F., Moscovitz, P., Hodge, J.W., Adams, J.P., 1976.Age related Ca–Mg content and strength in turkey tendon.Calcified Tissue Res. 19, 331–336.

Lieber, R.L., Leonard, M.E., Brown, C.G., Trestik, C.L., 1991.Frog semitendinosus tendon load strain properties duringpassive loading. Am. J. Physiol. 261, C86–C92.

Lieber, R.L., Leonard, M.E., Brown-Maupin, C.G., 2000.Effects of muscle contraction on the load-strain propertiesof frog aponeurosis and tendon. Cells Tissues Organs 166,48–54.

Liem, K.F., Summers, A.P., 1999. Gross anatomy and func-tional morphology of the muscles. In: Hamlett, W.(Ed.),Biology of Elasmobranchs. Johns Hopkins University Press,Baltimore,.

Liem, K.F., Walker, W.F., Bemis, W.E., Grande, L., 2000.Functional Anatomy of the Vertebrates: An EvolutionaryPerspective. Fort Wort, Harcourt College.

Long, J.A., 1995. The Rise of Fishes: 500 million Years ofEvolution. Johns Hopkins University Press, Baltimore.

Maisano, J.A., 2001. A survey of state of ossification inneonatal squamates. Herpetol. Monogr. pp. 135–157.

Martin, R.B., Burr, D.B., Sharkey, N.A., 1998. Skeletal TissueMechanics. Springer, New York.

McHenry, M.J., 2001. Mechanisms of helical swimming:asymmetries in the morphology, movement and mechanicsof larvae of the ascidianDistaplia occidentalis. J. Exp. Biol.204, 2959–2973.

Mohamed, F.M., Foutz, T.L., Rowland, G.N., Villegas, P.,1995. Biomechanical properties of the gastrocnemius tendonin broilers experimentally infected with avian reovirus.Trans. ASAE 38, 1893–1899.

Moss, M.L., Moss-Salentijn, L., 1983. Vertebrate cartilages.In: Hall, B.K. (Ed.), Cartilage. Academic Press, New York,pp. 1–24.

Niklas, K.J., 1992. Plant Biomechanics: An EngineeringApproach to Plant Form and Function. University of Chi-cago Press, Chicago.

Olson, W.M., 2000. Phylogeny, ontogeny, and function: Extras-keletal bones in the tendons and joints ofHymenochirusboettgeri (Amphibia: Anura: Pipidae). Zool., Anal. ComplexSyst. 103, 15–24.

Padian, K., 1997. Encyclopedia of Dinosaurs. Academic Press,San Diego.

Patterson, C., 1977. Cartilage bones, dermal bones andmembrane bones, or the exoskeleton versus the endoskele-ton. In: Andrews, Miles and Walker,(Eds.), Problems inVertebrate Evolution. Linnean Society Symposium Seriesnumber 4.

Person, P., 1983. Invertebrate cartilages. In: Hall, B.K.(Ed.),Cartilage. Academic Press, New York, pp. 31–58.

Pike, A.V.L., Ker, R.F., Alexander, R.M., 2000. The develop-ment of fatigue quality in high- and low-stressed tendons ofsheep(Ovis aries). J. Exp. Biol. 203, 2187–2193.

Pollock, C.M., Shadwick, R.E., 1994a. Allometry of muscle,tendon, and elastic energy storage capacity in mammals.Am. J. Physiol. 266, R1022–R1031.

Pollock, M., Shadwick, R.E., 1994b. Relationship betweenbody mass and biomechanical properties of limb tensions inadult mammals. Am. J. Physiol. 266, R1016–R1021.

Riegel, J.A., 1978. Factors affecting glomerular function inthe Pacific hagfishEptatretus stouti (Lockington). J. Exp.Biol. 73, 261–277.

Riemersma, D.J., Schamhardt, H.C., 1982. The cryo-jaw, aclamp designed for in vitro rheology studies of horse digitalflexor tendons. J. Biomech. 15, 619–620.

Rowe, R.W.D., 1985. The structure of the rat tail tendon.Connect. Tissue Res. 14, 9–20.

Ruppert, E.E., 1991. Cephalochordata(Acrania). In: Harrison,F.W., Ruppert, E.E.(Eds.), Microscopic Anatomy of Inver-tebrates. Wiley-Liss, New York, pp. 349–504.

Sarin, V.K., Erickson, G.M., Giori, N.J., Bergman, A.G., Carter,D.R., 1999. Coincident development of sesamoid bones andclues to their evolution. Anat. Rec. 257, 174–180.

Shadwick, R.E., Rapoport, H.S., Fenger, J.M., 2002. Structureand function of the tuna tail tendons. Comp. Biochem.Physiol. 133A, 1109–1125.

Shu, D.G., Luo, H.L., Morris, S.C., et al., 1999. LowerCambrian vertebrates from south China. Nature(Lond.)402, 42–46.

Smith, M.M., Hall, B.K., 1990. Development and evolutionaryorigins of vertebrate skeletogenic and odontogenic tissues.Biol. Rev. 65, 277–373.

Summers, A.P., 2000. Stiffening the stingray skeleton—aninvestigation of durophagy in myliobatid stingrays(Chon-drichthyes, Batoidea, Myliobatoidea). J. Morphol. 243,113–126.

Summers, A.P., Koob, T.J., Brainerd, E.L., 1998. Stingray jawsstrut their stuff. Nature(Lond.) 395, 450–451.

Sych, V.F., 2001. Evolutionary aspects of sesamoid ossifica-tions of leg muscle tendons in birds. J. Morphol. 248, 290.

Trestik, C.L., Lieber, R.L., 1993. Relationship between achillestendon mechanical properties and gastrocnemius musclefunction. J. Biomech. Eng. 115, 225–230.

Trueb, L., Pugener, L.A., Maglia, A.M., 2000. Ontogeny ofthe bizarre: An osteological description ofPipa pipa (Anura:Pipidae), with an account of skeletal development in thespecies. J. Morphol. 243, 75–104.

Vogel, F., Gemballa, S., 2000. Locomotory design of ‘cyclo-stome’ fishes: spatial arrangement and architecture ofmyosepta and lamellae. Acta Zool.(Stockholm) 81,267–283.

Vogel, F., Gemballa, S., 2001. Locomotory design of ‘cyclo-stome’ fishes: spatial arrangement and architecture ofmyosepta and lamellae. J. Morphol. 248, 296.

Vogel, K.G., Koob, T.J., 1989. Structural specializations intendons under compression. Int. Rev. Cytol. 115.

Vogel, S., 1988. Life’s Devices: The Physical World ofAnimals and Plants. Princeton University Press, Princeton,NJ.

1170 A.P. Summers, T.J. Koob / Comparative Biochemistry and Physiology Part A 133 (2002) 1159–1170

Weitbrecht, G., 2000. Funktionsanatomische Untersuchungenam Lokomotionsapparat von Branchiostoma lanceolatum(Pallas, 1774) (Cephalochordata: Branchiostomatidae). Tue-bingen, University of Tuebingen: 75 pp.

Weitbrecht, G.W., Gemballa, S., 2001. Arrangement and archi-tecture of myosepta in the lanceletBranchiostoma lanceo-latum. J. Morphol. 248, 299.

Westneat, M.W., Hoese, W., Pell, C.A., Wainwright, S.A.,1993. The horizontal septum: mechanisms of force transferin locomotion of scombrid fishes(Scombridae, Percifor-mes). J. Morphol. 217, 183–204.

Wren, T.A.L., Yerby, S.A., Beaupre, G.S., Carter, D.R., 2001.Mechanical properties of the human Achilles tendon. Clin.Biomech. 16, 245–251.