masseter electromyography during chewing in ring-tailed lemurs (lemur catta)

Masseter Electromyography During Chewing inRing-Tailed Lemurs (Lemur catta)

Christopher J. Vinyard,1* Christine E. Wall,2 Susan H. Williams,3 Kirk R. Johnson,2

and William L. Hylander2,4

1Department of Anatomy, Northeastern Ohio Universities College of Medicine, Rootstown, Ohio 442722Department of Biological Anthropology and Anatomy, Duke University Medical Center, Durham,North Carolina 277103Department of Biomedical Sciences, Ohio University College of Osteopathic Medicine, Athens, Ohio 457014Duke University Primate Center, Durham, North Carolina 27105

KEY WORDS jaw adductors; chewing; electromyography; symphyseal fusion

ABSTRACT We examined masseter recruitment andfiring patterns during chewing in four adult ring-tailedlemurs (Lemur catta), using electromyography (EMG).During chewing of tougher foods, the working-sidesuperficial masseter tends to show, on average, 1.7 timesmore scaled EMG activity than the balancing-side super-ficial masseter. The working-side deep masseter exhibits,on average, 2.4 times the scaled EMG activity of the bal-ancing-side deep masseter. The relatively larger activityin the working-side muscles suggests that ring-tailedlemurs recruit relatively less force from their balancing-side muscles during chewing. The superficial masseterworking-to-balancing-side (W/B) ratio for lemurs over-laps with W/B ratios from anthropoid primates. In con-trast, the lemur W/B ratio for the deep masseter is moresimilar to that of greater galagos, while both are signifi-cantly larger than W/B ratios of anthropoids. Becausering-tailed lemurs have unfused and hence presumablyweaker symphyses, these data are consistent with thesymphyseal fusion-muscle recruitment hypothesis stat-ing that symphyseal fusion in anthropoids providesincreased strength for resisting forces created by the bal-ancing-side jaw muscles during chewing. Among themasseter muscles of ring-tailed lemurs, the working-sidedeep masseter peaks first on average, followed in succes-sion by the balancing-side deep masseter, balancing-sidesuperficial masseter, and finally the working-side super-ficial masseter. Ring-tailed lemurs are similar to greater

galagos in that their balancing-side deep masseter peakswell before their working-side superficial masseter. Wesee the opposite pattern in anthropoids, where the bal-ancing-side deep masseter peaks, on average, after theworking-side superficial masseter. This late activity ofthe balancing-side deep masseter in anthropoids islinked to lateral-transverse bending, or wishboning, oftheir mandibular symphyses. Subsequently, the stressesincurred during wishboning are hypothesized to be aproximate reason for strengthening, and hence fusion, ofthe anthropoid symphysis. Thus, the absence of thismuscle-firing pattern in ring-tailed lemurs with theirweaker, unfused symphyses provides further correla-tional support for the symphyseal fusion late-acting bal-ancing-side deep masseter hypothesis linking wishboningand symphyseal strengthening in anthropoids. The earlypeak activity of the working-side deep masseter in ring-tailed lemurs is unlike galagos and most similar to thepattern seen in macaques and baboons. We hypothesizethat this early activity of the working-side deep massetermoves the lower jaw both laterally toward the workingside and vertically upward, to position it for the upcom-ing power stroke. From an evolutionary perspective, thedifferences in peak firing times for the working-side deepmasseter between ring-tailed lemurs and greater galagosindicate that deep masseter firing patterns are not con-served among strepsirrhines. Am J Phys Anthropol130:85–95, 2006. VVC 2005 Wiley-Liss, Inc.

Jaw-muscle activity patterns during chewing haveplayed an increasingly important role in furthering ourunderstanding of how the primate masticatory appara-tus functions (e.g., Hylander and Johnson 1985, 1994;Hylander et al., 1987, 2000, 2004, 2005; Lieberman andCrompton, 2000; Vinyard et al., 2005). Much of thiseffort has focused on comparing jaw-muscle activity pat-terns in anthropoids with their fused mandibular symphy-ses to strepsirrhines with their unfused symphyses. Untilrecently, strepsirrhines were represented solely by greatergalagos (Otolemur crassicaudatus and O. garnetti) (e.g.,Hylander et al., 2000). We report here on the electromyo-graphic (EMG) activity of the superficial and deep mass-eters during chewing in ring-tailed lemurs (Lemur catta).In addition to providing another strepsirrhine species withan unfused symphysis for comparison to anthropoids, thesedata will help us better understand masseter activity pat-terns in Malagasy lemurs. Prior to our work, jaw-muscleactivity patterns during chewing were virtually unknown

for this clade (Hylander et al., 2002, 2003, 2004, 2005;Vinyard et al., 2006).We concentrate on two aspects of jaw-muscle activity

in characterizing ring-tailed lemur masseter EMGs.First, we examine a hypothesis focusing on recruitmentpatterns or the relative amounts of muscle activity

Grant sponsor: NSF; Grant number: BCS-01-38565.

*Correspondence to: Christopher J. Vinyard, Department of Anat-omy, Northeastern Ohio Universities College of Medicine, 4209 St.,Rt. 44, PO Box 95, Rootstown, OH 44272.E-mail: [email protected]

Received 19 March 2004; accepted 4 March 2005.

DOI 10.1002/ajpa.20307Published online 12 December 2005 in Wiley InterScience

(www.interscience.wiley.com).

VVC 2005 WILEY-LISS, INC.

AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 130:85–95 (2006)

in the superficial and deep masseters during chewing.Second, we consider a hypothesis concerning the firingpatterns or timing of peak activity among the superficialand deep masseters. These two components of jaw-muscle activity are fundamental to understanding jaw-muscle function during chewing in primates (e.g.,Hylander et al., 1992, 2000, 2004, 2005; Hylander andJohnson, 1994; Weijs, 1994; Lieberman and Crompton,2000; Langenbach and van Eijden, 2001).

THE SYMPHYSEAL FUSION MUSCLE-RECRUITMENT HYPOTHESIS

Relative to corpus strain levels on the working- (orchewing-) side of the jaw, greater galagos exhibit muchlower magnitudes of bone strain on the balancing- (ornonchewing-) side corpus when compared to both owlmonkeys and macaques (Hylander, 1979a; Hylander et al.,1998). Similarly, galagos display relatively higher work-ing- to balancing-side (W/B) ratios for masseter EMGactivity during chewing as compared to anthropoids(Hylander et al., 2000). This correspondence in relativejaw-muscle activity and bone strain patterns on theworking vs. balancing sides of the jaw supports thehypothesis that anthropoids recruit relatively more forcethan galagos from their balancing-side jaw muscles dur-ing chewing (Hylander, 1977, 1979a,b; Hylander et al.,1998, 2000). Given that the greater galago symphysis islikely structurally weaker than the fused symphyses ofanthropoids (Beecher, 1977, 1979), these findings sup-port the hypothesis that symphyseal fusion in anthro-poids is a functional adaptation for strengthening thesymphysis to resist the increased internal forces createdby relatively greater recruitment of the balancing-sidejaw muscles during chewing and/or biting (Hylander,1975, 1979a,b; Beecher, 1977; Ravosa and Hylander,1994; Hylander et al., 1998, 2000, 2004, 2005).In studying the link between symphyseal fusion and

increased balancing-side masseter recruitment, Hylanderet al. (2000) tested whether differences in relative mas-seter recruitment in galagos vs. anthropoids occur in pri-marily 1) vertical, 2) transverse, or 3) a combination ofvertical and transverse directions. Galagos have largermean W/B ratios for both the superficial and deep masse-ters compared to anthropoids, but overlap in superficial-masseter W/B ratios across individual experiments. Thegalago deep-masseter W/B ratio, however, is larger thanthose for anthropoids across all experiments. Thus, dif-ferences in relative recruitment of balancing-side muscleswith a significant transverse direction of pull (i.e., thedeep masseters) show the strongest association with var-iation in symphyseal fusion (Hylander et al., 2000).

Prediction 1

The symphyseal fusion muscle-recruitment hypothesispredicts that ring-tailed lemurs with their unfusedsymphyses will have higher W/B ratios for their masse-ters than anthropoids. We operationalize this hypothesisby comparing ring-tailed lemur W/B ratios to the highestobserved W/B ratios among anthropoids sampled to date(Vinyard et al., 2005). Ring-tailed lemurs are predictedto have average W/B ratios that are significantly higherthan the baboon superficial-masseter ratio (1.9) and theowl monkey deep-masseter ratio (1.4) (Hylander et al.,2000). We are careful to indicate that support for thishypothesis does not require ring-tailed lemurs to havesignificantly higher W/B ratios than anthropoids for both

the superficial and deep masseters (Hylander et al.,2000; Vinyard et al., 2005). For example, if the ring-tailed lemur deep masseter is the only jaw-muscle W/Bratio that is significantly higher than that of anthro-poids, then this outcome supports the hypothesis linkingsymphyseal fusion to resisting increased levels of balanc-ing-side muscle force in a primarily transverse direction(Hylander et al., 2000). If, however, neither the superfi-cial- nor deep-masseter W/B ratios of ring-tailed lemursare significantly higher than anthropoids, then thishypothesis is rejected.

THE SYMPHYSEAL FUSION-LATE-ACTINGBALANCING-SIDE DEEP-MASSETER

HYPOTHESIS

The balancing-side deep masseter (BDM) in anthro-poids peaks later in the power stroke, at a time whenmost of the other jaw-closing muscles have alreadypeaked and are rapidly relaxing (e.g., after the peak ofthe working-side superficial masseter) (Hylander et al.,1987, 2000; Hylander and Johnson, 1994). This late peakactivity, along with the BDM’s large component of trans-verse pull and significant relative recruitment as com-pared to its working-side counterpart, indicates that it islikely a major contributor to lateral-transverse bending,or wishboning, of the anthropoid mandibular symphysis(Hylander and Johnson, 1994; Hylander et al., 2000).The early peak activity of the BDM in galagos, comparedto the late activity in anthropoids, led Hylander et al.(2000) to hypothesize that symphyseal fusion in anthro-poids is proximately related to the wishboning loadingregime and hence to the transversely oriented forces cre-ated by the late-acting balancing-side deep masseter dur-ing the power stroke.

Prediction 2

The symphyseal fusion late-acting BDM hypothesispredicts that ring-tailed lemurs with their unfused man-dibular symphysis will not exhibit a late peak activity oftheir BDM during chewing. This hypothesis will receivesupport if the BDM does not peak significantly laterthan the peak of the working-side superficial masseter.

MATERIALS AND METHODS

Subjects

We examined jaw-muscle activity patterns duringchewing in two male and two female adult ring-tailedlemurs (Lemur catta). All subjects were healthy adultswith complete dentitions, weighing between 1.75–2.75kg. We habituated individuals to accepting food prior torecording EMG data. We recorded EMG data from eachlemur during at least two and up to four separateexperiments. We present data from 13 separate experi-ments.

Jaw-muscle electromyography (EMG)

Electrodes and their placement. Our electrodeconstruction and general placement procedures followHylander and Johnson (1985, 1994) and Hylander et al.(2000). Prior to inserting electrodes, we sedated eachlemur with a combination of medetomidine (0.04 mg/kg),butorphanol (0.4 mg/kg), and midazolam (0.3 mg/kg)(Williams et al., 2003), and placed it in a restraining

86 C.J. VINYARD ET AL.

chair that allowed free movement of the head and neck.We used fine-wire, bipolar indwelling electrodes (nickel-chromium alloy, 0.05 mm diameter; California FineWire) to record activity in the masseter muscles. Toinsert an electrode, we placed the tip of the electrode ina 30-gauge needle and inserted it in the muscle until theneedle’s tip contacted bone. We then removed the needle,leaving the tip of the electrode in the muscle near thebone.We placed electrodes in the superficial masseter by

positioning the needle (with inserted electrode) midwaybetween the muscle’s anterior and posterior borders, andinserting it until it contacted the edge of the mandibularangle (Fig. 1). We placed deep-masseter electrodes byinserting the needle just below the zygomatic arch, mid-way between the mandibular condyle and the junction ofthe arch and postorbital bar (Fig. 1). The needle wasinserted at an approximately 308 downward angle rela-tive to the ramus until it contacted bone. We did not ver-ify electrode placement via dissection because subjectswere not killed at the end of an experiment.

Recording and quantifying EMGs. Followingelectrode implantation, we reversed the sedative andallowed the restrained animal to recover fully from seda-tion. Once an animal became fully alert, we began feed-ing it dried raisins and gummy bears (i.e., gelatincandy). We amplified and band pass-filtered (100–3,000Hz) EMG potentials from the masseters during chewing,and recorded them onto a 14-channel FM tape recorderat a rate of 15 inches/sec. We continued feeding thelemur until we had collected sufficient data or the ani-mal refused to eat any more. Following data collection,we removed the electrodes, freed the animal from therestraints, and returned it to its cage. All recoveriesfrom these procedures were uneventful.We converted the analog EMG data from selected

chewing sequences to digital data sampling at 10 kHz,and recorded the digital data to a computer using Lab-View. These digitized EMG data were filtered with a dig-ital Butterworth band-pass filter (100–3,000 Hz). We cal-culated the root-mean-square (rms) of each digitized rawEMG signal, using a 42-millisecond (msec) time constant

(see Fig. 2 in Hylander et al., 2000). We then identifiedthe largest peak rms value per electrode across all chewsin an experiment, and assigned it a value of 1.0. Theremaining smaller peaks were linearly rescaled to thisvalue for each electrode. These scaled values were usedto create W/B ratios (Hylander et al., 2000).

Jaw-muscle recruitment: W/B ratios. We deter-mined the relative working-to-balancing-side scaled ac-tivity level for the superficial and deep masseters, follow-ing the procedures outlined in Hylander et al. (2000,2005) and Vinyard et al. (2005). The working-to-balanc-ing-side (W/B) ratio measures the amount of scaledbalancing-side muscle activity relative to that muscle’sworking-side counterpart. For example, a W/B ratio of1.0 indicates equal amounts of relative muscle activity,while a W/B ratio of 3.0 indicates three times morescaled activity in the working-side acompared to the bal-ancing-side muscle.1

We calculated the mean and standard deviation of theW/B ratios for each masseter pair across all chews in anexperiment for dried raisins and gummy bears, respec-tively. Because W/B ratios tend to be right-skewed, W/Bratios were log10-transformed prior to calculating theseexperimental means. The log10 transformation helps nor-malize these data (Sokal and Rohlf, 1995), and results inour calculating geometric rather than arithmetic meansof W/B ratios.

Jaw-muscle firing patterns: timing. We exam-ined when the masseter muscles are active relative toeach other during a chewing cycle by comparing the tim-ing of their peak EMG activities. We compared the timeof each muscle’s rms EMG peak to the time of peakactivity in the working-side superficial masseter (WSM)(Hylander and Johnson, 1994; Hylander et al., 2000). Ifa peak EMG for a muscle preceded the WSM peak, thenwe reported the number of milliseconds this musclepeaked prior to it as a positive value. Alternatively, if amuscle’s peak EMG occurred after the peak of the WSM,the number of milliseconds was assigned a negativevalue. We also estimated the length of the chewing cycleduring rhythmic mastication as the time elapsedbetween the peaks of the WSM in consecutive chews (seeHylander and Johnson, 1994). This measure provides arelative means of interpreting the timing differencesamong these muscles. We calculated the mean timing ofeach muscle’s peak activity relative to the WSM acrossall chews in an experiment for both dried raisins andgummy bears, respectively.

Grand means and 95% confidence intervals.We report grand means for W/B ratios and jaw-musclepeak firing times during the chewing of dried raisinsand gummy bears, respectively, as the mean of theexperimental averages (n ¼ 9 for both foods). For eachW/B ratio, we averaged the log10 values from the nineexperimental means for each food, and reported theirrespective antilogs as the grand means.We estimated the 95% confidence interval (CI) for

these grand means using a bootstrapping approach(Efron, 1979; Manly, 1997). For each food type, we pooledthe nine experimental means for each variable (for W/B

Fig. 1. Lateral view of lemur skull and jaw muscles. Dotsdepict approximate placement of EMG electrodes in this study.We removed part of the superficial masseter here to show place-ment of the deep-masseter electrode. We placed the deep-mass-eter electrode by inserting it through the superficial masseterand into the deep masseter.

1We arbitrarily established a maximum W/B ratio at 10.0 for anysingle chew. We did this to reduce the likelihood that a few excep-tionally large values would inflate the mean W/B ratio (Hylanderet al., 2000).

87LEMUR MASSETER EMG DURING CHEWING

ratios, we used log10 means), and then resampled these val-ues with replacement to create 1,000 new samples eachwith nine entries. We then ranked the averages of these1,000 bootstrapped samples from smallest to largest. Thevalues at the 2.5th and 97.5th percentiles of this rankeddistribution were taken as the 95% CI for that grand mean.

Statistical comparisons for W/B ratios and timingof the balancing-side deep masseter

We used the 95% CI for the grand mean of each W/Bratio to test the symphyseal fusion muscle-recruitmenthypothesis (Vinyard et al., 2005). If a W/B ratio for

anthropoids falls below the 95% CI for ring-tailed lemurs,then this hypothesis is supported for that muscle pair. Weused a one-tailed sign test to determine whether the aver-age peak activity in the BDM occurs significantly later thanthe peak of the working-side superficial masseter (WSM).The symphyseal fusion late-acting balancing-side deepmasseter hypothesis is supported if the ring-tailed lemurBDM significantly precedes or does not differ from theWSM peak. Because previous research focused on EMGsduring mastication of harder/tougher foods, we used theexperimental means for masseter W/B ratios and peak tim-ing during chewing of the tougher gummy bears (n ¼ 9) intesting these hypotheses.

RESULTS

Masseter W/B ratios

The grand means for the superficial- and deep-masseterW/B ratios during chewing of dried raisins are 2.0 and 3.3,respectively (Table 1a). When chewing gummy bears, theaverage W/B ratios for the superficial and deep massetersdecrease to 1.7 and 2.4, respectively (Table 1b). Thus, onaverage, ring-tailed lemurs exhibit approximately 2–3 timesas much relative working-side masseter activity comparedto its balancing-side counterpart when chewing dried rai-sins (Fig. 2a) and gummy bears (Fig. 2b).The confidence intervals for the grand means of super-

ficial-masseter W/B ratios during chewing of dried rai-sins and gummy bears show significant overlap, despitethe average W/B ratio being smaller when chewinggummy bears (Table 1). In contrast, the confidence inter-

Fig. 2. Digitized rms EMG traces for masseter muscles dur-ing chewing. A: Masseter EMG during chewing of dried raisinsby Lemur 1. First five chews are on the right side, followed by fourleft-sided chews. For each muscle, EMG activity is higher on theworking side. B: Masseter EMG during chewing of gummy bearsby Lemur 2. First three chews are left-sided, while last five are onthe right side. Similar to Lemur 1, EMG activity is higher on theworking side. y-axis for each muscle, scaled EMG activity; x-axis,time in seconds. Vertical dashed line passes through peak of work-ing-side superficial masseter in each chewing cycle. LSM, leftsuperficial masseter; RSM, right superficial masseter; LDM, leftdeep masseter; RDM, right deep masseter.

Fig. 3. Bar graphs of the average percent EMG activity forthe working- and balancing-side superficial and deep massetersduring chewing of dried raisins and gummy bears. y-axis, per-centage of scaled EMG activity for specific muscle. White linesbelow the top of each bar indicate standard error of mean. WS,working-side superficial masseter; BS, balancing-side superficialmasseter; WD, working-side deep masseter; BD, balancing-sidedeep masseter.


vals for deep-masseter W/B ratios show little overlapbetween the two foods. While these data suggest thatchewing tougher foods (i.e., gummy bears over dried rai-sins) results in a relative increase in balancing-sidemasseter activity, this increase appears more substantialfor the BDM (Table 1).Comparing percent EMG activity during chewing of

dried raisins and gummy bears (Fig. 3) helps explain thedifferent magnitudes of change in W/B ratios betweenthe superficial and deep masseters. When lemurs chewgummy bears, they increase their average percent ofactivity in their working- and balancing-side superficialmasseters as well as their balancing-side deep masse-ters. This is consistent with the hypothesis that tougherfoods require greater jaw-muscle forces to break themdown. The percent activity of the working-side deepmasseter does not change appreciably between foodtypes (Fig. 3). We hypothesize that the similar percent-age of recruitment in this muscle when chewing dried

raisins vs. gummy bears relates to its primary functionof moving the jaw toward the working side to help alignthe teeth for the upcoming power stroke. Thus, the simi-lar percent of activity in the working-side deep masseteracross these two foods, coupled with the reduction inpercentage of balancing-side deep-masseter activitywhen chewing dried raisins, results in a greater changein W/B ratio between foods for the deep masseter com-pared to the superficial masseter (Table 1).

Masseter firing patterns

The working-side deep masseter (WDM), balancing-side superficial masseter (BSM), and balancing-side deepmasseter (BDM) each peak, on average, prior to the peakactivity of the working-side superficial masseter (WSM)(Table 2, Fig. 4). The WDM typically is the first jaw-clos-ing muscle to peak during the lemur chewing cycle (seealso Hylander et al., 2005). It peaks, on average, 37 msec

TABLE 1. Descriptive statistics for ring-tailed lemur masseter W/B EMG ratios1

Subject N

Superficial masseter Deep masseter

Mean SD log102 Mean SD log10

1a. Softer food: dried raisinLemur 1Exp. A, raisin 51 2.5 0.11 2.8 0.15Exp. B, raisin 63 1.9 0.11 3.3 0.18Exp. C, raisin 43 2.7 0.12 1.8 0.14Subject mean, raisin 2.3 0.08 2.6 0.14

Lemur 2Exp. A, raisin 40 2.2 0.16 2.7 0.17Exp. C, raisin 38 1.9 0.19 4.0 0.21Exp. D, raisin 16 1.9 0.20 3.2 0.12Subject mean, raisin 2.0 0.04 3.3 0.09

Lemur 3Exp. B, raisin 27 1.4 0.15 6.6 0.17Subject mean, raisin 1.4 6.6

Lemur 4Exp. A, raisin 19 2.0 0.20 3.1 0.10Exp. D, raisin 32 1.5 0.10 3.6 0.12Subject mean, raisin 1.7 0.09 3.3 0.05

Grand mean, raisin 2.0 0.09 3.3 0.1595% CI for grand mean, raisin 1.72–2.22 2.69–3.95

1b. Tougher food: gummy bearLemur 1Exp. C, gummy bear 51 3.1 0.12 1.8 0.12Subject mean, gummy bear 3.1 1.8

Lemur 2Exp. B, gummy bear 44 1.7 0.17 2.4 0.16Exp. C, gummy bear 49 1.5 0.23 3.1 0.21Exp. D, gummy bear 33 1.6 0.12 2.4 0.13Subject mean, gummy bear 1.6 0.03 2.6 0.06

Lemur 3Exp. A, gummy bear 43 1.6 0.12 2.7 0.17Exp. B, gummy bear 41 1.5 0.17 3.4 0.20Subject mean, gummy bear 1.6 0.02 3.0 0.07

Lemur 4Exp. A, gummy bear 44 2.1 0.16 2.4 0.13Exp. B, gummy bear 67 1.6 0.13 2.0 0.13Exp. C, gummy bear 37 1.3 0.14 1.9 0.13Subject mean, gummy bear 1.6 0.10 2.1 0.05

Grand mean, gummy bear 1.7 0.17 2.4 0.0995% CI for grand mean, gummy bear 1.51–2.01 2.11–2.75

1 Mean values indicate antilogged value of the mean W/B ratio for a given food type in a single experiment. Grand mean indicatesantilog of a mean of logged experimental means for a food. N, number of chewing cycles. SD log10 is standard deviation of log10-transformed ratios. 95% CI, confidence interval for grand mean, based on bootstrapping nine logged experimental means for driedraisins and gummy bears, respectively. Exp, experiment.2 Standard deviations were calculated only from the nine logged experimental means because of the difficulties in antilogging astandard deviation (e.g., Finney, 1941; Laurent, 1963).


prior to the WSM during the chewing of both dried rai-sins and gummy bears. The BDM is typically the nextmasseter muscle to peak, between 21–24 msec prior tothe WSM, depending on food type. The BSM peaks next,on average 14–17 msec prior to the WSM, depending onthe food. Individual lemur averages generally follow thistiming pattern for the masseters, although there is somevariation in the sequence of peak EMG activity acrossexperiments (Table 2). There is essentially little quanti-tative difference in the average peak timing among themasseter muscles when chewing dried raisins vs. gummybears. The confidence intervals for each muscle showconsiderable overlap for the two foods (Table 2).

Hypothesis tests for ring-tailed lemur EMG

The symphyseal fusion muscle-recruitment hypothe-sis predicts that the average W/B ratios for the masse-ter muscles of ring-tailed lemurs will be significantly

higher than those of anthropoid primates. We opera-tionalized this hypothesis by comparing the CI for eachmuscle pair in ring-tailed lemurs to the largest averageW/B ratio observed in anthropoids with comparableEMG data. The average W/B range for the superficialmasseter of ring-tailed lemurs during gummy bearchewing (1.7; CI ¼ 1.5–2.0) is lower than the baboonW/B ratio of 1.9. In contrast, the CI for the averagedeep masseter W/B ratio of lemurs (2.1–2.8) is higherthan the average W/B ratio observed among owl mon-keys (1.4). Based on these results, the ring-tailed lemurdata support the hypothesis linking symphyseal fusionto a relative increase in transversely directed forcesfrom the balancing-side deep masseter (Hylander et al.,2000).The symphyseal fusion late-acting BDM hypothesis

predicts that ring-tailed lemurs with their unfusedsymphyses will not exhibit peak firing of their BDM sig-nificantly later than the peak of the WSM. The sign test

TABLE 2. Timing differences in milliseconds between peak EMG activity in a jaw muscle and the peak of theworking-side superficial masseter in ring-tailed lemurs1

Subject N

Working deepmasseter

Balancingsuperficialmasseter

Balancing deepmasseter

Chewing cyclelength

Mean SD Mean SD Mean SD Mean SD

2a. Softer food: dried raisinLemur 1Exp. A, raisin 51 53 29.8 10 14.3 25 26.7 362 47.8Exp. B, raisin 63 30 19.0 11 21.1 22 22.9 367 53.6Exp. C, raisin 43 54 19.8 5 12.9 24 18.7 344 58.4Subject mean, raisin 46 13.6 9 3.2 24 1.5 358 12.1

Lemur 2Exp. A, raisin 40 58 29.0 30 32.0 17 31.3 320 56.6Exp. C, raisin 38 38 15.4 22 19.1 20 15.3 289 39.8Exp. D, raisin 16 29 20.4 24 10.4 17 13.5 257 49.9Subject mean, raisin 42 14.8 25 4.2 18 1.7 289 31.5

Lemur 3Exp. B, raisin 27 56 33.7 14 21.1 23 22.6 311 59.0Subject mean, raisin 56 14 23 311

Lemur 4Exp. A, raisin 19 13 7.7 21 13.4 26 13.1 240 35.6Exp. D, raisin 32 6 12.4 16 12.2 18 12.1 235 29.0Subject mean, raisin 10 4.9 19 3.5 22 5.7 238 3.5

Grand mean, raisin 37 19.4 17 7.9 21 3.5 303 50.695% CI for grand mean, raisin 25.4–48.2 12.8–21.6 19.3–23.2

2b. Tougher food: gummy bearLemur 1Exp. C, gummy bear 51 41 18.6 1 12.3 22 14.1 333 40.2Subject mean, gummy bear 41 1 22 333

Lemur 2Exp. B, gummy bear 44 40 21.2 5 17.5 23 18.2 339 66.5Exp. C, gummy bear 49 45 21.2 14 8.8 33 25.1 324 35.2Exp. D, gummy bear 33 39 17.1 12 10.8 26 19.2 326 38.9Subject mean, gummy bear 41 3.2 10 4.7 27 5.1 330 8.1

Lemur 3Exp. A, gummy bear 43 56 25.7 18 16.1 45 28.9 323 42.5Exp. B, gummy bear 41 37 19.7 11 16.4 19 29.0 331 64.3Subject mean, gummy bear 47 13.4 15 4.9 32 18.4 327 5.7

Lemur 4Exp. A, gummy bear 44 20 12.3 22 12.1 11 16.0 268 22.8Exp. B, gummy bear 67 26 16.7 27 14.3 20 18.3 281 38.9Exp. C, gummy bear 37 27 13.7 20 12.7 28 16.8 270 33.1Subject mean, gummy bear 24 3.8 23 3.6 17 12.2 273 7.0

Grand mean, gummy bear 37 11.0 14 8.3 24 11.1 310.6 28.895% CI for grand mean, gummy bear 30.8–43.6 9.3–19.1 17.7–31.2

1 Positive values indicate that the peak EMG activity of the muscle precedes peak EMG activity of the working-side superficialmasseter. Negative values indicate reverse condition. N, number of power strokes; SD, standard deviation. Exp, experiment.


indicates that, on average, the BDM peaks significantlyearlier than the WSM during the power stroke (Table 3).Consistent with this hypothesis, ring-tailed lemurs donot exhibit the late-acting BDM firing pattern.

DISCUSSION

The balancing-side deep masseter andsymphyseal fusion

One of our reasons for quantifying masseter activitypatterns during chewing in ring-tailed lemurs was toprovide a second strepsirrhine species with an unfusedsymphysis for comparison to anthropoid primates.2 Rela-tive to this aim, the ring-tailed lemur EMG plays a con-firmatory role in helping us understand how variation inmasseter EMG activity patterns between anthropoidsand greater galagos might relate to symphyseal fusion.Ring-tailed lemurs recruit their masseters more like

greater galagos than anthropoids (Table 4). Thereappears to be little distinction between anthropoids andstrepsirrhines in the superficial masseter W/B ratio dur-ing chewing. Indeed, lemurs have an average superficial-masseter W/B ratio that is intermediate among anthro-poids (Table 4a). Conversely, the average deep-masseterW/B ratio for ring-tailed lemurs is significantly largerthan those of anthropoids (Table 4a). The smaller deep-masseter W/B ratio of anthropoids suggests that these

primates may recruit relatively more force from theirbalancing-side deep masseter during chewing, comparedto these two strepsirrhines. Therefore, ring-tailed lemursalso support the symphyseal fusion muscle-recruitmenthypothesis linking fusion, as a means of strengtheningthe symphysis, to a relative increase in balancing-sidemuscle force oriented in a primarily transverse direction(Hylander et al., 2000). While the similar results for gal-agos and lemurs offer strong inferential support linkingsymphyseal fusion to transversely directed jaw-muscleforces, this conclusion assumes that these species areappropriate living models of chewing behaviors in theprimitive anthropoid with an unfused symphysis (seeVinyard et al., 2005).A primary difference in masseter firing patterns be-

tween anthropoids compared to greater galagos andring-tailed lemurs relates to the timing of peak activityof the BDM (Table 4b). In these strepsirrhines, the BDMpeaks before the WSM. The BDM peaks after the WSMin most anthropoids. The symphyseal fusion late-actingBDM hypothesis argues that fusion in anthropoids isproximately related to transversely oriented forces fromthe BDM that wishbone the symphysis during the powerstroke. Ring-tailed lemurs uphold this hypothesis, inthat their BDM peaks, on average, early in the powerstroke, much like greater galagos (Table 4b). In sum-mary, both the relative recruitment and firing pattern ofthe ring-tailed lemur BDM corroborate the pattern seenin greater galagos as compared to anthropoids. Thus,ring-tailed lemurs also support the hypothesis that sym-physeal fusion in anthropoids is a means of strengthen-ing the symphysis against transversely directed forcesgenerated by the late-acting BDM.

The early-peaking working-side deep masseter

The working-side deep masseter (WDM) in ring-tailedlemurs peaks, on average, much earlier than the superfi-cial masseter and temporalis muscles during the chewingcycle (Table 2; Hylander et al., 2005). This early peakactivity is similar to the WDM firing pattern in macaquesand baboons, but is unlike the WDM of greater galagos(Hylander et al., 2000) (Table 4b). The WDM is also thefirst masseter muscle to peak in owl monkeys and callitri-chids, but it peaks with the same-side temporalis in theseanimals, rather than significant earlier as seen in ring-tailed lemurs, macaques, and baboons (Hylander et al.,2000, 2004; Vinyard et al., 2006).We hypothesize that this early activity relates to mov-

ing the lower jaw laterally toward the chewing side andvertically upward during the closing stroke, to position itfor the upcoming power stroke (Fig. 5). The WDM’s rolein positioning the jaw might explain why food consis-tency has little effect on its average percent activity(Fig. 3). In other words, the jaw has to be properlyaligned for the upcoming power stroke, regardless of thefood being masticated. Figure 5 in Hylander et al. (2000)similarly shows that macaques and baboons exhibit littlechange in percent WDM activity when chewing hard/tough vs. soft foods. If our hypothesis is correct, then wewould expect to find that the WDM 1) contracts concen-trically (i.e., shortens), and 2) typically peaks before thestart of tooth-tooth or tooth-food-tooth contact (i.e., priorto the power stroke) in these species.Ravosa (1999) and Ravosa et al. (2000) argued that

the delayed balancing-side deep-masseter activity pat-tern in crown anthropoids, predominantly responsible

Fig. 4. Timing of percent EMG activity for the massetermuscles during chewing of gummy bears in ring-tailed lemurs(Lemur 2). Working-side deep masseter (WDM) peaks first, fol-lowed by the balancing-side deep masseter (BDM), balancing-sidesuperficial masseter (BSM), and finally the working-side superfi-cial masseter (WSM). y-axis, percent activity of jaw muscles risingfrom 25% to peak (100%) and then falling back to 25%. Time inmilliseconds is on x-axis, with the chewing cycle beginning at theleft and moving to the right as it progresses. Peak of WSM is arbi-trarily set to time zero as the reference muscle. Vertical line passesthrough the WSM peak.

2While ring-tailed lemurs maintain an unfused symphysis, theylikely have more connective tissue binding the two halves of theirmandibles together as compared to greater galagos (Beecher, 1977).Thus, ring-tailed lemurs are likely intermediate between greatergalagos and anthropoids in symphyseal strength and hence relativeload-resisting ability.


for producing the wishboning loading regime, might beevolutionarily linked to maintaining or increasing trans-verse jaw movements and/or transversely directedocclusal forces in this clade. Assuming that the wishbon-ing loading regime is linked to symphyseal fusion incrown anthropoids, the WDM activity pattern in ring-tailed lemurs offers an analogous test of the relativeimportance of transverse jaw movements vs. occlusalforces for symphyseal fusion. If the WDM functions pri-marily to move the jaw transversely (and vertically) dur-ing the closing stroke, then it appears that the unfusedsymphysis of ring-tailed lemurs is sufficiently con-structed to facilitate these transverse (and vertical) jawmovements. The implication of this observation is thatsymphyseal fusion in anthropoids is related to strength-ening the symphysis against increased transverselydirected occlusal forces during the power stroke, ratherthan solely facilitating increased transverse jaw move-ments.Our hypothesis that the WDM in ring-tailed lemurs,

macaques, and baboons primarily functions to move thejaw during the closing stroke also has implications forinterpreting deep-masseter W/B ratios. If the WDM israpidly shortening during fast closing (i.e., firing isotoni-cally) in these primates, then its magnitude of recruit-ment is more likely correlated with the speed of contrac-tion rather than its relative force. This suggests that the

deep-masseter W/B ratio may not correspond to the ratioof relative force produced by the working- and balanc-ing-side muscles (Hylander et al., 1992). Alternatively, inanimals where both the working- and balancing-sidedeep masseters are likely firing isometrically (or nearlyso), then the deep-masseter W/B ratio would tend to be abetter estimate of the ratio of relative muscle forces.The potential movement-related function of the WDM

suggests that the deep-masseter W/B ratio may be moreuseful as an indicator of relative muscle activity ratherthan as a relative muscle force estimate in some pri-mates. This use of the W/B ratio would likely beadequate for most comparative purposes, except that weare not sure all primates would exhibit this same rapidshortening of their WDM fibers during the closingstroke. In primates such as macaques, baboons, andring-tailed lemurs, where the WDM may be acting pri-marily to move the jaw during closing, WDM activitywould not increase significantly as food toughness and/orhardness increases. The W/B ratio would tend to belower in these animals compared to primates such asowl monkeys, callitrichids, and greater galagos, whereWDM activity would increase with increasing food hard-ness and/or toughness. This observation may helpexplain why macaques and baboons have deep-masseterW/B ratios of 1.0, while owl monkeys and callitrichidshave higher ratios of 1.3 (Hylander et al., 2000; Vinyard

TABLE 4. Summary comparisons of average masseter activity patterns among primates1

4a. W/B ratios Superficial masseter Deep masseter

SpeciesRing-tailed lemur2 1.7 2.4Greater galago 2.2 3.9Baboon 1.9 1.0Macaque 1.4 1.0Owl monkey 1.4 1.3Callitrichid 1.9 1.3

4b. Peak timing3 Working deep masseter Balancing superficial masseter Balancing deep masseter

SpeciesRing-tailed lemur 37 14 24Greater galago 11 21 23Baboon 47 17 �6Macaque 65 17 �20Owl monkey 13 �1 �11Callitrichid 10 �5 �12

1 Data for greater galagos (Otolemur crassicaudatus, O. garnetti), baboons (Papio anubis), macaques (Macaca fascicularis, M. fus-cata), and owl monkeys (Aotus trivirgatus) are from Hylander et al. (2000). Grand means for these species were calculated followingmethods outlined here. This accounts for slight differences in values reported here and in Hylander et al. (2000). Data for callitri-chids (Callithrix jacchus, Saguinus fuscicollis) are from Vinyard et al. (2006).2 We use average data during chewing of gummy bears for these comparisons, because data for these other animals reflect masseteractivity patterns while they were chewing the toughest foods these animals would eat.3 Positive values indicate that peak EMG activity of the muscle precedes peak EMG activity of the working-side superficial mass-eter. Negative values indicate the reverse condition.

TABLE 3. Sign tests comparing number of times that peak activity of a jaw muscle follows peak of the working superficial masseter1

Tougher food: gummy bear2

Working deep masseter Balancing superficial masseter Balancing deep masseter

Count 0.0 0.0 0.0P-value 0.004 0.004 0.004

1 Count indicates number of experimental means (out of nine) that the average peak firing time of a jaw muscle follows the peak ofthe working superficial masseter. P-values are based on sign tests comparing whether a muscle peaks significantly earlier or laterthan WSM. All P-values are significant at an adjusted Bonferroni a ¼ 0.05.2 Because previous interspecific comparisons focused on harder and tougher foods, we only consider the tougher gummy bears inthis hypothesis test. That being said, we observed some results for these statistical tests when lemurs chewed dried raisins.


et al., 2006). The implication of this potential functionaldifference in WDM activity is that the most relevant pri-mate suborder comparisons of deep-masseter W/B ratiosinvolve contrasting ring-tailed lemurs to macaques andbaboons (i.e., species potentially firing the WDM isotoni-cally), as well as greater galagos to owl monkeys and cal-litrichids (i.e., species primarily firing the WDM isomet-rically). In both of these comparisons, we find thatanthropoids with fused symphyses have significantlylower W/B ratios for their deep masseters as comparedto the respective strepsirrhine species.The early peak of the WDM in ring-tailed lemurs may

be derived among primates, based on comparisons tojaw-muscle activity patterns in treeshrews and greatergalagos (Hylander et al., 2000, 2005; Vinyard et al.,2005, 2006). On the other hand, the remaining jaw-closingmuscles of ring-tailed lemurs likely follow the primitiveprimate firing pattern reasonably closely (Vinyard et al.,2006). This early WDM firing pattern likely evolved

independently in Malagasy lemurs and Old World monkeys(Hylander et al., 2000, 2002, 2003; Vinyard et al., 2006). Itis also clear that the late-firing, balancing-side deep-masse-ter pattern also evolved independently in sifakas andanthropoids (Hylander et al., 2003). Thus, the timing ofpeak activity in the primate deep masseter appears to beeasily and routinely uncoupled from the superficial masse-ter and the other jaw-closing muscles throughout primateevolution (Vinyard et al., 2006). Ravosa et al. (2000) sug-gested that in living anthropoids, the activity patterns forthe superficial and deep masseters were uncoupled in thestem lineage leading to crown anthropoids. Our resultsindicate that this uncoupling may have occurred prior tothe crown anthropoid stem lineage.

Feeding ecology and masseter muscle activityduring chewing in ring-tailed lemurs

There is ample evidence that jaw-muscle activityvaries with the structural and mechanical properties ofthe foods that primates consume (e.g., Hiiemae, 1978;Hylander, 1979a; Hylander et al., 1992, 2000; Agrawalet al., 1998). Ring-tailed lemurs show slightly larger masse-ter W/B ratios when chewing dried raisins compared togummy bears. Given that gummy bears are tougherthan dried raisins (Williams et al., in press), it is notsurprising that these animals show relatively more bal-ancing-side masseter recruitment when chewing gummybears.Because jaw-muscle activity patterns may change with

food types, we should consider how foods fed to animalsin the laboratory relate to foods eaten in their naturalhabitats. This link is a prerequisite for making stronginferences about the biological roles of the jaw-musclesand hence their adaptations. Without this link, we areprimarily studying variation in the structural design ofthe masticatory apparatus among primates.

Fig. 5. Frontal view of a ring-tailed lemur (Lemur 4) chewingon the left side, with overlay of raw EMG traces from the deepmasseters. A: The animal is in beginning of closing stroke. Down-ward-pointing arrow marks approximate position of midline ofupper jaw. Upward-pointing arrow marks approximate position ofthe midline for lower jaw. Note that the lower jaw is positioned toanimal’s right of the upper jaw at this time in chewing cycle. Blackarrow indicates a position of EMG signal for working-side (i.e., left)deep masseter (WDM). There is minimal deep-masseter activity atthis point. Inset: Frontal-view diagram of the lower jaw’s orbit dur-ing the chewing cycle on left side. Vertical line indicates the mid-line of maxilla. Horizontal line indicates superiormost extent oflower jaw during chewing orbit. This superiormost position isreached during centric occlusion (i.e., at maximal intercuspation).Dot on horizontal line labeled LC indicates position of centric occlu-sion during left-sided chewing cycle. RC indicates same for right-sided chewing cycle. Dot indicated by arrow near bottom of orbitindicates approximate position of the lower jaw along orbit at thispoint in the chewing cycle. B: The animal is now nearing end ofclosing stroke. Upward-pointing arrow indicating position of thelower jaw midline is now slightly to the left of midline of animal’supper jaw. Black arrow indicates that WDM has already peaked(as seen in large blip just to the left of arrow) and is shutting off.On the trace below the WDM, it appears that the balancing-sidedeep masseter (BDM) is increasing in activity at this instance inthe chewing cycle. Inset: Dot indicated by arrow shows approxi-mate position of the lower jaw in its orbit at this point in the chew-ing cycle. In both A and B, EMG trace information to the left ofsolid black line was removed because it reflects previous chewingcycles not depicted in these two images.


Yamashita (2000, 2002, 2003) measured the toughnessof the ring-tailed lemur diet during long-term studies oftheir feeding ecology at Beza Mahafaly Special Reserve,Madagascar. Ring-tailed lemurs consumed fruits andleaves with an average toughness of approximately 500J/m2, with the middle 50% of these data ranging fromapproximately 300–1,000 J/m2 (see Fig. 6b in Yamashita,2002). Ripe kily fruits (Tamarindus indica) were thetoughest food consumed by ring-tailed lemurs, with anaverage toughness of 3,504 J/m2 (Yamashita, 2000). Weused a similar device to determine that the dried raisinsand gummy bears we fed subjects in the laboratory havean average toughness of 418 J/m2 (SD, 107) and 888 J/m2

(SD, 114), respectively (Williams et al., in press). Theseresults suggest that dried raisins and gummy bears fallwithin the middle 50% of toughness values for foods con-sumed by lemurs in their natural habit.This comparison suggests that we have captured jaw-

muscle activity for the average food toughness these ani-mals consume. Importantly, this means we have notapproached the upper limit of food toughness these animalseat in the wild, and hence may see different muscle-activitypatterns when lemurs consume these tougher foods. Basedon our EMG data, W/B ratios would likely decrease withthese tougher foods, while firing patterns may not changeappreciably. We recognize, however, that it is unclearwhether primate jaw muscles are adapted for chewing themost common foods in their diet and/or the most difficultfoods they consume. Finally, we are not suggesting thattoughness is the only physical parameter to be analyzed, orthat the foods we fed these animals are similar to foods thatring-tailed lemurs eat in the wild simply because they havesimilar toughness values. Our intent is to highlight theneed for further studies linking field research on primatediets and experimental research on the primate mastica-tory apparatus.

CONCLUSIONS

We examined masseter muscle activity during chewing inring-tailed lemurs (Lemur catta). Ring-tailed lemur mas-seters are significantly more active on the working side ofthe jaw as compared to the balancing side during chewing.Across experiments, the average superficial-masseter W/Bratios for ring-tailed lemurs overlap with those of anthro-poid primates, while average deep-masseter W/B ratios arealways larger than anthropoid W/B ratios. Because ring-tailed lemurs have unfused symphyses, this result is consis-tent with the hypothesis that symphyseal fusion in anthro-poid primates is a structural adaptation for resistingincreased, primarily transversely directed forces from thebalancing-side jaw muscles during chewing.On average, the working-side deep masseter (WDM) of

ring-tailed lemurs peaks first during the chewing cycle,followed by the balancing-side deep masseter (BDM), thebalancing-side superficial masseter (BSM), and lastly theworking-side superficial masseter (WSM). The very earlypeak of the WDM is likely a convergence with Old Worldmonkeys. We hypothesize that this early peak activity isfunctionally related to moving the jaw laterally towardthe working side, and vertically upward during the closingstroke. The BDM of ring-tailed lemurs is similar togreater galagos in peaking before the WSM on average.This pattern is different from that seen in nonhumananthropoids, who peak their BDM significantly after theWSM peak. Symphyseal fusion in anthropoids is hypothe-sized to be a structural adaptation for resisting wishbon-

ing forces, created in large part by this late-peaking BDM(Hylander et al., 2000). Ring-tailed lemurs support thishypothesis, given their unfused symphyses and early-peaking BDM.Relative masseter activity levels increased on the bal-

ancing side when we fed lemurs tougher foods. The mas-seter firing pattern, however, did not appear significantlyaltered between softer dried raisins vs. tougher gummybears. Both of these foods have toughness values that aresimilar to foods typically eaten by ring-tailed lemurs in thewild. Even though several other food properties remain tobe analyzed and compared, this similarity represents animportant first step in linking feeding ecology to jaw-muscle activity patterns measured in the laboratory.

ACKNOWLEDGMENTS

We thank D. Brewer, K. Glenn, B. Hess, J. Ives, C.Williams, and the Duke University Primate Center forhelping us collect these data. M.J. Ravosa, N. Yama-shita, and two anonymous reviewers offered critical com-ments that improved the manuscript. This is Duke Uni-versity Primate Center Publication no. 783.

LITERATURE CITED

Agrawal KR, Lucas PW, Bruce IC, Prinz JF. 1998. Food proper-ties that influence neuromuscular activity during human mas-tication. J Dent Res 77:1931–1938.

Beecher RM. 1977. Function and fusion of the mandibularsymphysis. Am J Phys Anthropol 47:325–336.

Beecher RM. 1979. Functional significance of the mandibularsymphysis. J Morphol 159:117–130.

Efron B. 1979. Bootstrap methods: another look at the jack-knife. Ann Stat 7:1–26.

Finney DJ. 1941. On the distribution of a variate whose logarithmis normally distributed. J R Stat Soc Lond 222:309–368.

Hiiemae KM. 1978. Mammalian mastication: a review of theactivity of the jaw muscles and the movements they producein chewing. In: Butler PM, Joysey KA, editors. Development,function and evolution of teeth. London: Academic Press. p361–398.

Hylander WL. 1975. The human mandible: lever or link? Am JPhys Anthropol 43:227–242.

Hylander WL. 1977. In vivo bone strain in the mandible of Gal-ago crassicaudatus. Am J Phys Anthropol 46:309–326.

Hylander WL. 1979a. Mandibular function in Galago crassicau-datus and Macaca fascicularis: an in vivo approach to stressanalysis. J Morphol 159:253–296.

Hylander WL. 1979b. The functional significance in primatemandibular form. J Morphol 160:223–240.

Hylander WL, Johnson KR. 1985. Temporalis and massetermuscle function during incision in macaques and humans. IntJ Primatol 6:289–322.

Hylander WL, Johnson KR. 1994. Jaw muscle function andwishboning of the mandible during mastication in macaquesand baboons. Am J Phys Anthropol 94:523–547.

Hylander WL, Johnson KR, Crompton AW. 1987. Loading pat-terns and jaw movements during mastication in Macaca fasci-cularis: a bone-strain, electromyographic, and cineradio-graphic analysis. Am J Phys Anthropol 72:287–314.

Hylander WL, Johnson KR, Crompton AW. 1992. Muscle forcerecruitment and biomechanical modeling: an analysis of mas-seter muscle function during mastication in Macaca fascicula-ris. Am J Phys Anthropol 88:365–387.

Hylander WL, Ravosa MJ, Ross CF, Johnson KR. 1998. Mandib-ular corpus strain in primates: further evidence for a func-tional link between symphyseal fusion and jaw-adductormuscle force. Am J Phys Anthropol 107:257–271.


Hylander WL, Ravosa MJ, Ross CF, Wall CE, Johnson KR.2000. Symphyseal fusion and jaw-adductor muscle force: anEMG study. Am J Phys Anthropol 112:469–492.

Hylander WL, Vinyard CJ, Wall CE, Williams SH, Johnson KR.2002. Recruitment and firing patterns of jaw muscles duringmastication in ring-tailed lemurs. Am J Phys Anthropol[Suppl] 34:88.

Hylander WL, Vinyard CJ, Wall CE, Williams SH, Johnson KR.2003. Convergence of the ‘‘wishboning’’ jaw-muscle activitypattern in anthropoids and strepsirrhines: the recruitmentand firing of jaw muscles in Propithecus verreauxi. Am J PhysAnthropol [Suppl] 36:120.

Hylander WL, Vinyard CJ, Ravosa MJ, Ross CF, Wall CE, John-son KJ. 2004. Jaw adductor force and symphyseal fusion. In:Anapol F, German R, Jablonski N. editors. Shaping primateevolution: form, function and behavior. Cambridge: CambridgeUniversity Press. p 229–257.

Hylander WL, Wall CE, Vinyard CJ, Ross CJ, Ravosa MJ, Wil-liams SH, Johnson KR. 2005. Temporalis function in anthro-poids and strepsirrhines: an EMG study. Am J Phys Anthropol128:35–56.

Langenbach GEJ, van Eijden TMGJ. 2001. Mammalian feedingmotor patterns. Am Zool 41:1338–1351.

Laurent AG. 1963. The lognormal distribution and the transla-tion method: description and estimation problems. J Am StatAssoc 58:231–235.

Lieberman DE, Crompton AW. 2000. Why fuse the mandibular symph-ysis? A comparative analysis. Am J Phys Anthropol 112:517–540.

Manly BFJ. 1997. Randomization, bootstrap and Monte Carlomethods in biology. London: Chapman and Hall.

Ravosa MJ. 1999. Anthropoid origins and the modern symphy-sis. Folia Primatol (Basel) 70:65–78.

Ravosa MJ, Hylander WL. 1994. Function and fusion of themandibular symphysis in primates: stiffness or strength? In:Fleagle JG, Kay RF, editors. Anthropoid origins. New York:Plenum Press. p 447–468.

Ravosa MJ, Vinyard CJ, Gagnon M, Islam SA. 2000. Evolutionof anthropoid jaw loading and kinematic patterns. Am J PhysAnthropol 112:493–516.

Sokal RR, Rohlf FJ. 1995. Biometry. New York: W.H. Freemanand Co.

Vinyard CJ, Williams SH, Wall CE, Johnson KR, Hylander WL.2005. Jaw-muscle electromyography during chewing in Belanger’streeshrews (Tupaia belangeri). Am J Phys Anthropol 127:26–45.

Vinyard CJ, Ravosa MJ, Williams SH, Wall CE, Johnson KR,Hylander WL. 2006. Jaw-muscle function and the origin ofprimates. In: Ravosa MJ, Dagosto M, editor. Primate originsand adaptations: a multidisciplinary perspective. New York:Kluwer Press. In press.

Weijs WA. 1994. Evolutionary approach of masticatory motorpatterns in mammals. In: Bels VL, Chardon M, Vandewalle P,editors. Biomechanics of feeding in vertebrates. Berlin:Springer-Verlag. p 282–320.

Williams CV, Glenn KM, Levine JF, Horne WA. 2003. Compari-son of the efficacy and cardiorespiratory effects of medetomi-dine-based anesthetic protocols in ring-tailed lemurs (Lemurcatta). J Zoo Wildlife Med 34:163–170.

Williams SH, Wright BW, Den Truong V, Daubert CR, VinyardCJ. No date. The mechanical properties of foods used inexperimental studies of the primate masticatory apparatus.Am J Primatol (in press).

Yamashita N. 2000. Mechanical thresholds as a criterion forfood selection in two prosimian primate species. In: SpatzH-C, Speck T, editors. Proceedings of the 3rd Plant Biome-chanics Conference, Freiburg-Badenweiler. Stuttgart: ThiemeVerlag. p 590–595.

Yamashita N. 2002. Diets of two lemur species in differentmicrohabitats in Beza Mahafaly Special Reserve, Madagascar.Int J Primatol 23:1025–1051.

Yamashita N. 2003. Food procurement and tooth use in twosympatric lemur species. Am J Phys Anthropol 121:125–133.


masseter electromyography during chewing in ring-tailed lemurs (lemur catta)

Documents