Activities of Pongid Thigh Muscles during Bipedal Behavior RUSSELL H. TUTTLE,' JOHN V. BASMAJIAN * AND HIDEMI ISHIDA3 I Department of Anthropology and Committee on Evolutionary Biology, The University of Chicago, Chicago, Illinois 60637, McMaster University and Rehabilitation Centre, Chedoke Hospitals, Hamilton, Ontario, Canada L8N 3L6; and Faculty of Human Sciences, Osaka University, Suita, Osaka, Japan
K E Y WORDS Apes Electromyography Positional behavior . Bipedalism
.
Thighmuscles
.
ABSTRACT Electromyographic recordings were taken from 1 2 thigh muscles (or major parts of them) in a gorilla, from 6 thigh muscles in a chimpanzee, and from 2 thigh muscles in an orangutan as they engaged in bipedal positional behavior, including stance, reaching overhead, lunging, leaping and walking. In the African apes, symmetric bipedal stances with hindlimb flexure were accompanied by notable EMG activity in many of the hamstring, quadriceps, and adductor muscles. If weight shifted eccentrically onto one hindlimb, EMG potentials generally increased to or remained at moderate and high levels. Our studies on the gluteal (Tuttle et al., '78) and thigh muscles of African apes partly confirm Kummer's ('75) prediction that considerable gluteal and hamstring activity would be required in order for them to stand bipedally with flexed hip and knee joints. The gorilla's thigh muscles exhibited considerable EMG activity during the stance phase and remarkably little activity during the swing phase of bipedal steps. The activity patterns of most thigh and gluteal muscles (Tuttle et al., '78) in the African apes are much more similar to those of bipedal gibbons than to their counterparts in man. The bipedal locomotor cycles of human subjects are accompanied by many more biphasic and triphasic EMG patterns in the thigh muscles than the locomotor cycles of other anthropoid primates are. The evolutionary anthropological significance of these findings should become clearer when they are complemented by EMG studies on human running, arboreal bipedalism and vertical climbing in apes, and central pattern generation in man and apes. The pongid apes, particularly the African ones, evince remarkably close molecular and karyological similarities to man (King and Wilson, '75; Sarich and Cronin, '76; Miller, '77). This indicates close genetic affinities among extant Pongidae and Hominidae and especially between Pan (including P. gorilla) and Homo. Yet in many morphological features, Pan and Homo are strikingly dissimilar. A principal task for comparative morphologists and evolutionary anthropologists is to explain the physical transformations which probably occurred between ancestral and modern Hominoidea during a relatively short period of time. One of t h e most persistent AM. J. PHYS. ANTHROP.(1979) 50: 123-136
classic puzzles is the evolution of hominid bipedalism. A burgeoning fossil record of osteological features relating to the problem of bipedalism exists for the period between 3 myBP and t h e present time. But there is virtually no direct evidence representing the period between 15and 3 myBP wherein most authorities believe that habitual bipedalism must have been established in the hominid lineage (Tuttle et al., '78). Many primates engage in facultative bipedalism. But we have only rudimentary and chiefly conjectural intelligence about how this bipedal behavior is maintained and the extent to which counterparts of it might rep-
123
124
R. H. TUTTLE, J. V. BASMAJIAN AND H. ISHIDA
resent hypothetical substrates for the development of habitual bipedalism as seen in Homo sapiens. Here we present a description of electromyographic (EMG) activities in the thigh muscles of bipedal pongid apes (table 1). This information is complementary t o that which we reported on the gluteal muscles of t b same subjects (Tuttle and Basmajian, '75; Tuttle et al., '75, '781, and the hip and thigh muscles of other anthropoid primates (Ishida et al., '75; Ishida et al., '78). SUBJECTS AND METHODS
The subjects in this study were a 4.75- to 5.5-year-old female lowland gorilla (Pan gorilla), a 5 - to 5.75-year-old male common chimpanzee fPun troglodytes), and a 7-year-old female Sumatran orangutan fPongo p y g maeusi. Indwelling fine-wire bipolar electrodes (Basmajian, '74, pp. 35-39) were used according t o procedures which have been described in previous papers (Tuttle and Basmajian, '74a,b; Tuttle et al., '75). The subjects could move freely in the testing area. They had the opportunity to stand bipedally on level and inclined surfaces (figs. la,c, Za, 3a), t o reach overhead for food incentives (figs. 2c,d, 3a-c1 and a trapeze (figs. la,b, 3c), and to walk bipedally at various speeds on the floor (figs. Id, 2b) and a ramp. RESULTS
General levels of EMG activity in individual thigh muscles during a variety of bipedal positional behaviors are listed in table 1. The nature of EMG recordings and the small number of subjects preclude sophisticated quantification and statistical renderings of our data. Data on the gorilla are fairly comprehensive. The project ended before we could complete studies on the thigh muscles of Pun troglodytes and Pongo pygmaeus. The orangutan did not lunge, leap or walk bipedally while thigh muscles were being recorded. Unless otherwise mentioned, all descriptions are based on data from fully alert subjects. Stance The gorilla and chimpanzee evinced flexure of the hip and knee joints (between 120' and 140")as they stood with their hindlimbs symmetrically loaded (fig. 2a). Their thighs were variably rotated laterally and abducted and their feet were plantigrade. The orangutan rarely stood quietly bipedally, unassisted by
hand contact with adjacent surfaces or objects. She generally showed less flexure of the hip and knee joints and more lateral rotation of the thighs than the African apes did. Weight fell predominently on the lateral aspects of her feet (fig. 3a). In the gorilla, bipedal stance was accompanied by notable EMG activity. Low or moderate potentials very commonly or always occurred in the rectus femoris, vastus lateralis, vastus intermedius, biceps femoris, semimembranosus, adductor magnus proprius and adductor longus muscles (figs. 4-6). The vastus medialis and long head of the adductor magnus muscles exhibited nil or low potentials and the semitendinosus and gracilis muscles were usually silent during symmetric stance (figs. 5, 6). Silence sometimes occurred in the vastus intermedius muscle as the subject stood with her knees flexed. During stances (generally hand assisted) with the knees extended, low and moderate potentials occurred in the vastus intermedius muscle. Somewhat higher potentials occurred in the proximal one-third of the adductor magnus proprius and gracilis muscles if the subject stood with her thighs widely abducted. In the chimpanzee, bipedal stance was accompanied by low potentials in the rectus femoris and vastus lateralis muscles, low and moderate potentials in the long head of the biceps femoris muscle, silence or occasionally low potentials in the adductor magnus proprius muscle, and silence in the vastus medialis and semimembranosus muscles (fig. 7). The major differences between the two African apes are the somewhat higher EMG activi t y of the biceps femoris muscle in the chimpanzee versus higher EMG activity in the semimembranosus muscle of the gorilla. In the orangutan, the rectus femoris muscle generally exhibited moderate or high potentials (and occasionally low potentials) as she stood with full alignment of the leg and thigh, often while lightly touching adjacent surfaces or objects. The long head of the biceps femoris muscle was silent as she stood with full extension of the hip and knee joints. However, when she stood with knee flexure, high and moderate potentials occurred in the long head of the biceps femoris muscle. As the African apes shifted additional weight onto the right hindlimb, EMG activity increased to (or remained at) moderate or high levels in the right rectus femoris, vastus lateralis, vastus intermedius, biceps femoris, semi-
G
c
2
2
1
Semitendinosus
Semimembranosus
2
G
+,+ +,+ + +
+,++
O,+,++
+
O,[+,++l
+,+ + 0
++,+++ OS+I
O,+,+
0, +
O,+,++
++
++
+,I++I
0
++.+++ ++
+,+ +
0
+ +,l++ + I
0
O,+,++
0
[+,++I
0
-
0
0, +
O.+.+
+ +
0
++
0
+,++
0
-
-
-
-
~
-
0
0
-
+++ ++ 0,+,+ +
+
0
-
0
+++
+++ ++ ++,+++
-
0
-
-
++,+++
~~
+++
-
0
+ +.+ + + -
-
-
++
Stance phase
~
-
+,+ +
+,++.++ +
0
-
++
4
-,
0
~~
0, +
0
+, low
~~
O,+,[+ +I
0
0
0
n
-
0
0
-
0
0
+,+ +
-
0
0
0
no data, 0.silence;
+,+++
0
+,+ +
+t
+,+ +
0
0
O,+,++,+++
~~
++
-
0
Swing phase
Walking
+ + , + + + +,++,+++ +,++,+++ +,++
Leap
+ +,+ + +
-
~
++
-
-
-
+,++
++
++
0, +
0
0
++
0
+,+ t
+,++,+++
o,++,+++
I+.+
++ +,++
0,+,[0,+,++1
+,++
++,+++
+.+ + +1 ++,+++ +,++,+++
+,+ +,+ + + +,++,+++
0
[0,+,++1
+,+ +
++,[+,+++I
++,+++,lo1
Lunge
+,+ + +,++ +,+ +
+,++
Digitigrade reach
Symbols: ', proxima1 one.third of thigh; t, middle one-third of thigh; G, gorilla. C, chimpanzee; 0.orangutan; N , number of 60 120minute experiments; S, subject; EMG; + C , moderate EMG; + + +, high EMG; [ I , indicate relatively rare occurrences or mainly while the subject was groggy.
Gracilis
c
G
1
1
G
It
Adductor longus
G
2*
Adductor magnus vrovi-ius
G
1
Adductor magnus (long head)
G
1
O,+,++
o,++,+++
0
1
Biceps femoris (short head)
G
+,+ t
c
2
Biceps femoris (lone head)
+,lo,+]
+
G
0
2
3
O,+,+
c
O,+,[O,+,++l
+
+,+ t
G
1
c
G
1
1
G
++,+++
+,++,+++
0
1
3
-
+
c
+,++,l+++l
+,++,+++
+,++,I01
G
Plantigrade reach
Eccentric loading stance
3
S
Quiet symmetric stance
2
N
Vastusintermedius
Vastus medialis
Vastus lateralis
Rectus femoris
Muacle
Actiuities ofpongid thigh muscles during bipedal behavior
TABLE 1
cn
til
CL
0
-
126
R. H. TU’ITLE, J. V. BASMAJIAN AND H. ISHIDA
a. b. c. d.
Fig. 1 Gorilla female. Standing on the apex of a ramp with her eccentrically loaded right hindlimb extended. Standing on a platform with her eccentrically loaded right hindlimb flexed. Standing with the left foot plantigrade and the right foot “on points” but bearing little weight. Walking bipedally with her flexed left hindlimb in mid stance phase and her right foot approaching toe off.
EMG OF PONGID THIGH MUSCLES AND BIPEDALISM
a. b. c. d.
Fig. 2 Chimpanzee male. Standing bipedally with his feet plantigrade and flexure of both hindlimbs. Walking bipedally while reaching overhead. Reaching for a food sill with his extended right hindlimb eccentrically loaded. Reaching for a suspended food incentive with his left foot gripping the edge of a ramp and with his right foot "on points."
Fig. 3 Orangutan female. a. Reaching overhead with flexure of her hindlimbs and weight resting mostly on her heels and lateral soles. b. Reaching for incentives on a sill with her hindlimba extended. c. Standing on the apex of a ramp with her eccentrically loaded right hindlimb nearly fully extended.
127
128
R. H. TUTTLE, J . V. BASMAJIAN AND H. ISHIDA
Fig. 4 EMG potentials from the right rectus femoris (a), vastus lateralis ib) and ischiofemoralis (c) muscles of a bipedal gorilla as she shifted weight eccentrically onto her flexed right hindlimb and pivoted clockwise on her right foot (secs. 1-21; during the stance phase of a bipedal step forward (secs. 2-3);during a relatively quite symmetrical stance with hip and knee flexure (secs. 4-8); during the stance phase of a slow forward step with the foot quasi-plantigrade while reaching overhead (secs. 8-9);and as she stood “on points” while manipulating a trapeze bar (secs. 10-12).(t, time in seconds. The vertical line between secs. 3 and 4 indicates that the illustration is based on discrete episodes during the recording session .)
membranosus (gorilla only), adductor magnus and adductor longus muscles. The vastus medialis muscle of the chimpanzee and the semitendinosus and gracilis muscles of the gorilla usually remained silent when the right hindlimb was eccentrically loaded during bipedal stance. In the orangutan, the rectus femoris muscle exhibited moderate and high potentials and the long head of the biceps femoris muscle was silent during eccentric loadings of the fully extended hindlimb.
Reaching The subjects used a variety of hindlimb postures when reaching overhead for the trapeze and food incentives. The hip and knee joints could be flexed (figs. lb, 2c,d, 3a,c) or extended (figs. la, 3b). The feet could be plantigrade (with heel contact: figs. la,b, 2c, 3a,c).
quasi-plantigrade (with dorsiflexion at the mid-tarsal joint so that the posterior heel was free of the substrate), or digitigrade (with the ankle markedly volarflexed and the digits extended so that only their distal pads were in contact with the floor: figs. 2d, 3b) which we have termed on points (Tuttle et al., ’75: p, 257). In table 1,plantigrade reach also includes incidents wherein the African apes used quasi-plantigrade foot postures. The hindlimbs were either somewhat flexed or fully extended during plantigrade reaches. The hips and knee joints were fully extended or overextended (i.e., beyond 180’) during high reaches with both feet on points. In the gorilla, plantigrade reaching was accompanied by low or moderate potentials in the rectus femoris, vastus lateralis, vastus intermedius, biceps femoris, semimembranosus, adductor magnus, and adductor longus mus-
129
EMG OF PONGID THIGH MUSCLES AND BIPEDALISM
C
1
2
3
4
5
6
7
8
9
sec. Fig. 5 EMG potentials from the right gracilis la), proximal adductor nlagnua proprius (bl, mid adductor magnus proprius (c). and short head of the biceps femoris !d) muscles of a bipedal gorilla. Secs. 1-2:a bipedal steps forward in which a acts during swing phase and b, c and d act during stance phase. Sec. 2-- vertical line t h e subject steps forward and stands bipedally with hip and knee flexure and hoth feet plantigrade; a acts mainly during the swing of the foot and b, c and d act while t h e hindlimb is loaded. Secs. 8-9: a leap to the trapeze from a crouched posture in which t h e thighs were widely abducted. (t and vertical line as in fig. 4.)
cles. High potentials occurred quite rarely in the reaching gorilla's rectus femoris muscle and then only when her knee was flexed. Once her semimembranosus muscle exhibited high potentials as she stumbled heavily onto the flexed right hindlimb while reaching overhead. Low and nil potentials occurred in the vastus medialis muscle. The semitendinosus and gracilis muscles were always silent during plantigrade and digitigrade reaching. As the gorilla rose t o the on points position (fig. 41,EMG levels commonly increased in the rectus femoris muscle. Sometimes while the
knee was fully extended, moderate and high potentials persisted; but at other times EMG activity decreased. The vastus medialis muscle also exhibited increased EMG activity when the gorilla reached in the digitigrade posture with full extension of her knees. There were no consistent differences in the EMG levels of the vastus lateralis, short head of the biceps femoris, semimembranosus and long head of the adductor magnus muscles. Sometimes EMG levels decreased in the vastus lateralis muscle after the gorilla had grabbed the food sill or trapeze. EMG levels often de-
1
2
3
4
5
6
7
8
9
10
11
12
sec. Fig. 6 EMG potentials from the right long head of the adductor magnus (a), gracilis (b), and adductor longus (c) muscles of a bipedal gorilla. Secs. 1-6and 10-12: a and c act during the stance phase of bipedal steps as the subject walks about while reaching overhead. Secs. 7-9:the subject lunges for the trapeze uligh potentials just after sec. 7) and then stands bipedally with her thighs widely abducted.
creased dramatically in the adductor magnus proprius muscle, especially in its middle onethird, as the gorilla achieved full extension of the knee during digitigrade reaching. In the chimpanzee, both plantigrade and digitigrade reaching were accompanied by low and moderate potentials in the rectus femoris and vastus lateralis muscles. Most reaching occurred beneath the food sill where the chimpanzee frequently hoisted himself bimanuall y or by the right forelimb alone instead of remaining bipedal on the floor. Thus, some of the observed decreases in EMG levels in the rectus femoris and vastus lateralis muscles as he achieved full extension of the hindlimbs may have occurred because he had begun to hoist himself, thereby lessening the load on his feet. Low, high, or most commonly moderate potentials occurred in the long head of the biceps femoris muscle during plantigrade reaches. Since (when not hoisting) he preferred to reach for the food sill with his left hand, often his right hindlimb was eccentrically loaded (fig. 2c). The chimpanzee's semimembranosus muscle was silent during all reaches. His adductor magnus proprius muscle was usually
silent during reaching though occasionally low or moderate potentials occurred during plantigrade reaching. In the orangutan, plantigrade reaching with the hip and knee joints fully extended was accompanied by moderate and high potentials in the rectus femoris muscle and silence in the long head of the biceps femoris muscle. The latter muscle was also silent during digitigrade reaching in which the hindlimb was fully extended.
Lunging and leaping Lunges (wherein the distal portions of the feet remained on the floor) and leaps (wherein the feet rose completely off the floor) were executed by the African apes as they attempted to grab the trapeze and food incentives on a sill or wall. The chimpanzee lunged and leaped much more frequently than the gorilla did. In the gorilla, moderate potentials occurred in the rectus femoris and short head of the biceps femoris muscles during lunges and moderate and high potentials accompanied leaps (fig. 5). The semitendinosus and gracilis
EMG OF PONGID THIGH MUSCLES AND BIPEDALISM
131
C
1
2
3
4
5
6
8
1
see. Fig. 7 EMG potentials from the right gluteus medius (a), rectus femoris (b), and adductor magnus proprius (c) muscles of a bipedal chimpanzee. The subject rises from a quadrupedal to a bipedal posture (sec. 1) and takes two steps (sees. 1-31. a and b act during the stance phase of each step. c acts as the subject rises (aec. 1) and then during the swing phase of each step (prior to secs. 2 and 3). Sec. 3- vertical line: subject stands with his thighs widely abducted and with his right knuckled hand propped on a wall. Secs. 7-8: vigorous leap from a crouched knuckle-walking posture to a trapeze bar. Note that potentials begin in a (associated with erecting the trunk by hip extension) before b commences activity (associated with propulsive knee extension). (t and vertical line as in fig.4.)
muscles were silent during lunges and the gracilis muscle was silent during leaps. The gorilla only lunged once while we were recording the long head of the adductor magnus muscle. It produced low potentials. The proximal one-third of the adductor magnus proprius muscle exhibited high potentials and its middle one-third exhibited moderate potentials during all lunges and leaps (fig. 5). Leaps and lunges were accompanied by high potentials in the adductor longus muscle (fig. 6 ) . In the chimpanzee, the rectus femoris muscle exhibited potentials ranging from low to moderate during lunges. Leaps were accompanied commonly by high or moderate potentials and rarely by low potentials in the rectus femoris muscle (fig. 7). The higher potentials were especially characteristic of leaps for the trapeze where, unlike leaping below the food sill, he could not augment the propulsive ef-
fort of the hindlimbs by hoisting. Lunges were accompanied by moderate potentials in the vastus lateralis muscle. Moderate potentials also occurred during most leaps; but high potentials were exhibited by the vastus lateralis muscle during especially vigorous leaps for the trapeze. Per contra the vastus medialis was silent during leaps. The semimembranosus muscle was also typically silent during leaping and lunging. When the chimpanzee lunged for the trapeze or food sill, the long head of the biceps femoris muscle exhibited moderate or high potentials. All leaps for the trapeze were accompanied by high potentials in the long head of the biceps femoris muscle. The adductor magnus proprius muscle exhibited quite variable EMG levels, ranging from nil to moderate potentials, during lunges. Leaps were characteristically accompanied by moderate or high potentials in the
132
R. H. TUTTLE, J. V. BASMAJIAN AND H. ISHIDA
chimpanzee's adductor magnus proprius muscle (fig. 7 ) . Bipedal walking The two African apes exhibited similar patterns of bipedal walking. They did not differ remarkably from descriptions of other bipedal chimpanzees and gorillas (Elftman, '44; Tuttle, '70; Jenkins, '72; Ishida et al., '75). Throughout each locomotive cycle, their knee and hip joints were flexed and the knee did not pass posteriorly beyond the ipsilateral hip joint. There was no evidence of pelvic tilt of a hominid sort. The subjects alternately leaned laterally over each supportive foot producing a waddling gait (Tuttle and Basmajian, '75; Tuttle et al., '75, '78). Many thigh muscles were quite active during the stance phase of the bipedal locomotor cycle in the African apes. In all instances, the activity consisted of uniphasic bursts of potentials which occurred predominantly during load bearing and propulsive segments of the stance phase. In the gorilla, moderate or high potentials usually or always occurred in the rectus femoris, vastus lateralis, vastus intermedius, proximal one-third of the adductor magnus proprius, and the adductor longus muscles during the stance phases of bipedal steps (figs. 4-6). Low potentials occasionally occurred in the rectus femoris and vastus intermedius muscles during some steps, especially when the gorilla touched a wall or other supportive objects. The short head of the biceps femoris muscle exhibited potentials ranging from nil t o high, with low and moderate ones being most common, during the stance phase of bipedal steps (fig. 5 ) . Once as the gorilla took several steps backwards, only inconsequential single potentials occurred in the short head of the biceps femoris muscle during the stance phase of each step. The long head of the adductor magnus and the middle one-third of the adductor magnus proprius muscles usually exhibited low and occasionally moderate potentials during the stance phase of bipedal steps (figs. 5, 6). The gorilla's semitendinosus and gracilis muscles were silent during the stance phase of bipedal steps (figs. 5 , 6). All thigh muscles which we recorded in the gorilla, except the gracilis, were silent during the swing phase of bipedal locomotor cycles. The gracilis muscle exhibited low and inconsequential single potentials during the swing phase of some steps (figs. 5,6).
In the chimpanzee, the rectus femoris and vastus lateralis muscles usually exhibited moderate and, in some instances, low potentials during the stance phase of bipedal steps (fig. 7). The long head of the biceps femoris muscle also exhibited low and moderate potentials during the stance phase of bipedal steps. Generally EMG activity was lower in the biceps femoris muscle during bipedal walking than during lunging, leaping, reaching and standing with the right hindlimb eccentrically loaded. The semimembranosus and adductor magnus proprius muscles were silent during the stance phase of all bipedal locomotor cycles (fig. 7). Two of the six thigh muscles recorded in the chimpanzee exhibited EMG activity during the swing phase of bipedal steps. Brief bursts of moderate and low potentials occurred in the vastus medialis muscle during swing phase, just prior to foot contact. The vastus medialis muscle was silent during all of stance phase. Thus it acted quite discretely from the vastus lateralis and rectus femoris muscles. Although the swing phases of most steps were accompanied by silence in the adductor magnus proprius muscle, a few were accompanied by low potentials and moderate potentials occurred during one step (fig. 7 ) . DISCUSSION
Comparisons with other EMGStudies on non-human primates Using surface electrodes, Ishida et al. ('75) recorded EMG activity in the vastus lateralis and long head of the biceps femoris muscles in a 7.5- t o 8-year-old female chimpanzee which had been trained over a 6-year period to walk bipedally on a level platform. Both muscles acted during the stance phase of bipedal steps. This is basically coincident with our results on a younger male chimpanzee, which had not experienced an extended period of bipedal training. Further, the actions of the vastus lateralis muscle in Pan gorilla and Pan troglodytes and of the long head of the biceps femoris muscle in P. troglodytes are not notably different from actions of their counterparts in Hylobates agilis, Macaca fuscata, Papio harnadryas, and Ateles geoffroyi during bipedal walking (Ishida et al., '75). In all species, they are active during the stance phase of each step (fig. 5 in Ishida et al., '75). Because the methods of recording and analyzing data differ between our study and that of Ishida et al. ('751, we cannot ascribe major functional impor-
EMG OF PONGID THIGH MUSCLES AND BIPEDALISM
tance to apparent dissimilarities in levels of EMG activity between them, With indwelling fine-wire bipolar electrodes, Ishida et al. ('78) recorded the activities of many thigh and hip muscles in a juvenile male HyEobates agiEis and a juvenile female Hylobates lar as they walked bipedally on a level platform. The hamstring, quadriceps femoris, sartorius, gracilis, adductor longus, adductor magnus, and gluteal muscles were quite active and the iliopsoas muscle exhibited minor activity during the stance phase of bipedal steps. The iliopsoas and rectus femoris muscles of Hylobates were active during most of (especially mid) swing phase and the gracilis and adductor longus muscles were active during release of the foot and early swing phase. Ishida et al. ('75) concluded that the latter two muscles initiate the swing phase by slightly raising the thigh. Then the iliopsoas and rectus femoris muscles carry the limb forward. The rectus femoris muscle also extends the knee joint to reposition the foot during the swing phase. The vastus lateralis muscles of both African apes, the vastus intermedius and adductor magnus muscles of the gorilla, and the long head of the biceps femoris muscle of the chimpanzee act basically like their counterparts in gibbons during bipedal progression. As in gibbons, the rectus femoris muscles of the African apes and the adductor longus muscle of the gorilla are active during the stance phase. But they differ from their counterparts in the gibbons in that they are silent during the swing phase of bipedal steps. By remaining silent during the stance phase of bipedal steps, the vastus medialis, semimembranosus and adductor magnus proprius muscles of the chimpanzee and the semitendinosus and gracilis muscles of the gorilla differ strikingly from their counterparts in the gibbons. This might be attributed t o the fact that the gibbons usually moved rapidly whereas the African apes progressed more slowly. It is also possible that the African apes walk in a manner which loads the medial thigh relatively less than in the gibbons. The gorilla resembles the gibbons somewhat in actions of the gracilis muscle during the swing phase of bipedal steps. The activity of the vastus medialis muscle during the late swing phase in the chimpanzee contrasts his bipedal walking with the bipedalism of the gorilla and two gibbons. EMG patterns from the gluteal muscles (gluteus maximus proprius, ischiofemoralis,
133
gluteus medius and gluteus minimus) of the two gibbons are basically similar to those of the gorilla and chimpanzee, viz. notable activity during the stance phase and silence during the swing phase of bipedal steps (Ishida et al., '78; Tuttle and Basmajian, '75; Tuttle et al., '75, '78). Functional implications Our studies of EMG activity in the thigh and gluteal muscles in African apes during bipedal stance partly confirm Kummer's ('75: pp. 285-2861 prediction t h a t considerable gluteal and hamstring activity would be required in order for a chimpanzee to stand bipedally with flexure of the hips and knees. The gluteus maximus proprius, ischiofemoralis, gluteus medius (fig. 7) and gluteus minimus muscles generally exhibit low or nil potentials during quiescent bipedal stance, though sometimes higher potentials occur, especially in the ischiofemoralis and gluteus minimus muscles (Tuttle et al., '78). This modest activity of the gluteal muscles to maintain hip extension (figs. 4,7) is especially complemented by low and moderate activity in the biceps femoris muscle (on the lateral aspect of the thigh) in both African apes and by the semimembranosus muscle (on the medial aspect of the thigh) in the gorilla. The other thigh muscles which exhibit notable potentials during bipedal stance probably act mainly t o sustain the partially flexed knee joint and the laterally rotated and abducted thigh against gravitational forces. As bipedal African apes fully extend their knee and hip joints, thereby reducing the moment of the extensor muscles (Kummer, '75: p. 2861, some muscles act quite prominently, especially when the on points posture is employed. Of the muscles which we recorded, this is particularly true for the gluteus maximus proprius (Tuttle et al., '75, '781, rectus femoris, vastus lateralis, vastus intermedius, and perhaps the biceps femoris. Not surprisingly, in the African apes many hip and knee extensor muscles are very active during the propulsive thrust of lunges and leaps. Often the subjects leap from crouched postures in which the thighs are flexed, abducted and laterally rotated and the knees are markedly flexed. Of the muscles which we recorded in the chimpanzee and gorilla during leaps and lunges, the gluteus maximus proprius, ischiofemoralis, gluteus medius (fig. 7), gluteus minimum (Tuttle et al., '781, and
134
R.H. TU'ITLE, J . V . BASMAJIAN AND H. ISHIDA
biceps femoris (long head) are major hip extensors; the adductor magnus and adductor longus are major adductors of the hip joint; and the rectus femoris and vastus lateralis are major knee extensors. Elftman ('44: p. 70) suggested that during the mid stance phase of bipedal walking in the chimpanzee the hip extensors must be active in order to restore the elevation of the body while concurrently increasing its forward velocity. This prediction is confirmed by our observations of notable activity in the gluteus maximus proprius, ischiofemoralis (fig. 4), gluteus medius (fig. 71, gluteus minimus (Tuttle et al., '781, and biceps femoris (long head) muscles during the stance phases of most bipedal steps by t h e African apes. In the gorilla, the long head of the adductor magnus and perhaps the adductor magnus propius and adductor longus muscles might also contribute to hip extension during the stance phase of bipedal steps. Comparisons with m a n
The EMG patterns exhibited by thigh and hip muscles in man and apes are generally strikingly dissimilar during most bipedal behavior and especially during bipedal progression. Human quiescent bipedal stance is remarkably economical in that no thigh and probably only a few hip muscles are active (Basmajian, '74; Carlsoo, '72; Okada, '72, '75). Integrity of the extended hip and knee joints is largely premised on osseoligamentous mechanisms in plantigrade standing humans. Further, Okada ('72, '75) reported that even when human subjects rise onto the "balls" of their feet, the gluteus maximus, rectus femoris, vastus lateralis, vastus medialis, semitendinosus and biceps femoris muscles remain silent or evince very low EMG activity. Per contra, quite high EMG activity occurs regularly in the rectus femoris muscle of the orangutan when i t stands bipedally with full alignment of the hindlimb. In plantigrade human subjects, flexure of the knee and hip joints to positions approximating those of bipedal pongid apes is accompanied by major increases in the activity of the quadriceps femoris and biceps femoris muscles, modest increases in the activity of the gluteus maximus muscle and little change in the semitendinosus muscle (Okada, '72, '75). Like human stance and unlike pongid bipedalism, human bipedal walking is accom-
panied by remarkably little EMG activity in the thigh muscles (Basmajian, '74: p. 308). During bipedal walking, human subjects also differ strikingly from pongid apes and other anthropoid primates in the much greater frequency of biphasic and triphasic EMG patterns evinced by human thigh muscles (Basmajian, '74: pp. 323-325; Carlsoo, '72: p. 117) versus uniphasic ones in the non-human subjects. In human subjects, the quadriceps femoris and hamstring muscles act from mid to late swing phase through heel strike and early stance (initial double support) phase and again during mid stance (single support) phase. The human adductor muscles act during late swing phase through initial double support phase and again from second double support phase into early swing phase (Carlsoo, '72: p. 117). The evolutionary problem of hominid bipedalism: what d o we need i n order toproceed ? Commentators on the evolution of hominid bipedalism often emphasize (and indeed marvel a t ) the achievement of balance and the economy of muscular activity in the human lower limb. Our findings are in general agreement with this perspective. Yet because human kinesiological studies are focussed on sexually mature subjects we are ignorant about how and when this condition is developed ontogentically. Additional studies on ad libitum human running, sustained terrestrial bipedal locomotion in apes and arboreal walking and running by apes are also needed before we can begin to sketch the many small steps in hypothetical evolutionary sequences which might have culminated in bipedal Homo sapiens. Experimental studies on arboreal bipedalism and vertical climbing in apes might prove to be particularly informative re possible preadaptations for hominid terrestrial bipedalism. The neurological bases for complex biphasic and triphasic activities of human lower limb muscles should also be explored in detail. Is there central pattern generation related to bipedalism in Homo sapiens? To what extent is this homologous with the neurological substrates for bipedal or other positional behavior in non-human primates and especially with neuromuscular mechanisms in man's closest living relatives, the apes? Finally, we should mention that whatever hypotheses we might formulate on the basis of
EMG OF PONGID THIGH MUSCLES AND BIPEDALISM
kinesiological studies of living primates, we must have the skeletons of ante- and early hominids in order to test their validity. It now seems more likely than ever t h a t such materials will be available before or a t the time that the challenging kinesiological studies mentioned herein above are implemented. ACKNOWLEDGMENTS
This study was supported mainly by NSF Grants GS-3209 and SOC75-02478 and by a Public Health Service Research Career Development Award (1-K04-GM16347-01) from the National Institutes of Health. Supplementary support was provided by NIH Grant RR00165 to the Yerkes Regional Primate Research Center and the Marian and Adolph Lichtstern Fund of t h e University of Chicago. We are especially thankful for the assistance of Doctor G. H. Bourne, J. Malone, E. Regenos, J. Perry, R. Pollard, S. Lee, R. Mathis, J. Roberts, Doctor M. Keeling, Doctor M. Vitti, J. Hudson, K. Barnes and J. Rike. This paper is dedicated to the memory of Mrs. Almas Benjamin, who died on March 12, 1978. She loved primates and endeavored to impart an appreciation of them to pupils a t the Christopher House Day Care Center in Chicago, Illinois. LITERATUHE CITED Basmajian, J. V. 1974 Muscles Alive. Their Functions Revealed by Electromyography. Third ed. The Williams & Wilkins Co., Baltimore. CarlsM, S. 1972 How Man Moves. Kinesiological Methods and Studies. Wm. Heinemann Ltd., London. Elftman, H. 1944 The bipedal walking of t h e chimpanzee. J. Mamm., 25: 67-71. Ishida, H., T. Kimura and M. Okada 1975 Patterns of bipedal walking in anthropoid primates. In: Symp. 5th Cong. Int’l. Primat. SOC.(1974). S. Kondo, M. Kawai, A. Ehara and S. Kawamura, eds. Japan Science Press, Tokyo, pp. 287-301. Ishida, H., M. Okada, R. H. Tuttle and T. Kimura 1978 Ac-
135
tivities of hindlimb muscles in bipedal gibbons. In: Recent Advances in Primatology. Vol. 3. D. 3. Chivers and K. A. Joysey, eds. Academic Press, London, pp. 459-462. Jenkins, F. A,, Jr. 1972 Chimpanzee bipedalisrn: cineradiographic analysis and implications for the evolution of gait. Science, 178: 871-879. King, M.-C., and A. C. Wilson 1975 Evolution a t two levels in humans and chimpanzees. Science, 188: 107-116. Kummer, B. K. F. 1975 Functional adaptation to posture in t h e pelvis of man and other primates. In: Primate Functional Morphology and Evolution. R. H. Tuttle, ed. Mouton, The Hague, pp. 281-290. Miller, D. A. 1977 Evolution of primate chromosomes. Science, 198: 1116-1124. Okada, M. 1972 An electromyographic estimation of the relative muscular load in different human postures. J. Human Ergo]., 1: 75-93. 1975 Quantitative studies on t h e bearing of t h e anti-gravity muscles in human postures with special references to electromyographic estimation of t h e postural muscle load. J. Fac. Sci., Univ. Tokyo, Sec. V, Vol. IV, pp. 471-530. Sarich, V. M., and J. E. Cronin 1976 Molecular systematics of the primates. In: Molecular Anthropology. M. Goodman and R. E. Tashian, eds. Plenum, New York, pp. 141-170. Tuttle, R. H. 1970 Postural, propulsive and prehensile capabilities in the cheiridia of chimpanzees and other great apes. In: The Chimpanzee, Vol. 2. G. H. Bourne, ed. Karger, Basel, pp. 167-253. Tuttle, R. H., and J. V. Basmajian 1974a Electromyography of brachial muscles in Pun gorilla and hominoid evolution. Am. J. Phys. Anthrop., 41: 71-90. 1974b Electromyography of forearm musculature in gorilla and problems related to knuckle-walking. In: Primate Locomotion. F. A. Jenkins, Jr.,ed. Academic Press, New York, pp. 293-347. 1975 Electromyography of Pan gorilla: a n experimental approach to t h e problem of hominization. In: Symp 5th Cong. Int’l. Primat. Soc. (1974). S. Kondo, M. Kawai, A Ehara and S. Kawamura, eds. Japan Science Press, Tokyo, pp. 303-314. Tuttle, R. H., J. V. Basmajian and H. Ishida 1975 Electromyography of the glUteUR maximus muscle in gorilla and t h e evolution of bipedalism. In: Primate Functional Morphology and Evolution. R. H . Tuttle, ed. Mouton, The Hague, pp. 253-269. 1978 Electromyography of pongid gluteal muscles and hominid evolution. In: Recent Advances in Primatology. Vol. 3. D. J. Chivers and K. A. Joysey, eds. Academic Press, London, pp. 463-468.
K E Y WORDS Apes Electromyography Positional behavior . Bipedalism
.
Thighmuscles
.
ABSTRACT Electromyographic recordings were taken from 1 2 thigh muscles (or major parts of them) in a gorilla, from 6 thigh muscles in a chimpanzee, and from 2 thigh muscles in an orangutan as they engaged in bipedal positional behavior, including stance, reaching overhead, lunging, leaping and walking. In the African apes, symmetric bipedal stances with hindlimb flexure were accompanied by notable EMG activity in many of the hamstring, quadriceps, and adductor muscles. If weight shifted eccentrically onto one hindlimb, EMG potentials generally increased to or remained at moderate and high levels. Our studies on the gluteal (Tuttle et al., '78) and thigh muscles of African apes partly confirm Kummer's ('75) prediction that considerable gluteal and hamstring activity would be required in order for them to stand bipedally with flexed hip and knee joints. The gorilla's thigh muscles exhibited considerable EMG activity during the stance phase and remarkably little activity during the swing phase of bipedal steps. The activity patterns of most thigh and gluteal muscles (Tuttle et al., '78) in the African apes are much more similar to those of bipedal gibbons than to their counterparts in man. The bipedal locomotor cycles of human subjects are accompanied by many more biphasic and triphasic EMG patterns in the thigh muscles than the locomotor cycles of other anthropoid primates are. The evolutionary anthropological significance of these findings should become clearer when they are complemented by EMG studies on human running, arboreal bipedalism and vertical climbing in apes, and central pattern generation in man and apes. The pongid apes, particularly the African ones, evince remarkably close molecular and karyological similarities to man (King and Wilson, '75; Sarich and Cronin, '76; Miller, '77). This indicates close genetic affinities among extant Pongidae and Hominidae and especially between Pan (including P. gorilla) and Homo. Yet in many morphological features, Pan and Homo are strikingly dissimilar. A principal task for comparative morphologists and evolutionary anthropologists is to explain the physical transformations which probably occurred between ancestral and modern Hominoidea during a relatively short period of time. One of t h e most persistent AM. J. PHYS. ANTHROP.(1979) 50: 123-136
classic puzzles is the evolution of hominid bipedalism. A burgeoning fossil record of osteological features relating to the problem of bipedalism exists for the period between 3 myBP and t h e present time. But there is virtually no direct evidence representing the period between 15and 3 myBP wherein most authorities believe that habitual bipedalism must have been established in the hominid lineage (Tuttle et al., '78). Many primates engage in facultative bipedalism. But we have only rudimentary and chiefly conjectural intelligence about how this bipedal behavior is maintained and the extent to which counterparts of it might rep-
123
124
R. H. TUTTLE, J. V. BASMAJIAN AND H. ISHIDA
resent hypothetical substrates for the development of habitual bipedalism as seen in Homo sapiens. Here we present a description of electromyographic (EMG) activities in the thigh muscles of bipedal pongid apes (table 1). This information is complementary t o that which we reported on the gluteal muscles of t b same subjects (Tuttle and Basmajian, '75; Tuttle et al., '75, '781, and the hip and thigh muscles of other anthropoid primates (Ishida et al., '75; Ishida et al., '78). SUBJECTS AND METHODS
The subjects in this study were a 4.75- to 5.5-year-old female lowland gorilla (Pan gorilla), a 5 - to 5.75-year-old male common chimpanzee fPun troglodytes), and a 7-year-old female Sumatran orangutan fPongo p y g maeusi. Indwelling fine-wire bipolar electrodes (Basmajian, '74, pp. 35-39) were used according t o procedures which have been described in previous papers (Tuttle and Basmajian, '74a,b; Tuttle et al., '75). The subjects could move freely in the testing area. They had the opportunity to stand bipedally on level and inclined surfaces (figs. la,c, Za, 3a), t o reach overhead for food incentives (figs. 2c,d, 3a-c1 and a trapeze (figs. la,b, 3c), and to walk bipedally at various speeds on the floor (figs. Id, 2b) and a ramp. RESULTS
General levels of EMG activity in individual thigh muscles during a variety of bipedal positional behaviors are listed in table 1. The nature of EMG recordings and the small number of subjects preclude sophisticated quantification and statistical renderings of our data. Data on the gorilla are fairly comprehensive. The project ended before we could complete studies on the thigh muscles of Pun troglodytes and Pongo pygmaeus. The orangutan did not lunge, leap or walk bipedally while thigh muscles were being recorded. Unless otherwise mentioned, all descriptions are based on data from fully alert subjects. Stance The gorilla and chimpanzee evinced flexure of the hip and knee joints (between 120' and 140")as they stood with their hindlimbs symmetrically loaded (fig. 2a). Their thighs were variably rotated laterally and abducted and their feet were plantigrade. The orangutan rarely stood quietly bipedally, unassisted by
hand contact with adjacent surfaces or objects. She generally showed less flexure of the hip and knee joints and more lateral rotation of the thighs than the African apes did. Weight fell predominently on the lateral aspects of her feet (fig. 3a). In the gorilla, bipedal stance was accompanied by notable EMG activity. Low or moderate potentials very commonly or always occurred in the rectus femoris, vastus lateralis, vastus intermedius, biceps femoris, semimembranosus, adductor magnus proprius and adductor longus muscles (figs. 4-6). The vastus medialis and long head of the adductor magnus muscles exhibited nil or low potentials and the semitendinosus and gracilis muscles were usually silent during symmetric stance (figs. 5, 6). Silence sometimes occurred in the vastus intermedius muscle as the subject stood with her knees flexed. During stances (generally hand assisted) with the knees extended, low and moderate potentials occurred in the vastus intermedius muscle. Somewhat higher potentials occurred in the proximal one-third of the adductor magnus proprius and gracilis muscles if the subject stood with her thighs widely abducted. In the chimpanzee, bipedal stance was accompanied by low potentials in the rectus femoris and vastus lateralis muscles, low and moderate potentials in the long head of the biceps femoris muscle, silence or occasionally low potentials in the adductor magnus proprius muscle, and silence in the vastus medialis and semimembranosus muscles (fig. 7). The major differences between the two African apes are the somewhat higher EMG activi t y of the biceps femoris muscle in the chimpanzee versus higher EMG activity in the semimembranosus muscle of the gorilla. In the orangutan, the rectus femoris muscle generally exhibited moderate or high potentials (and occasionally low potentials) as she stood with full alignment of the leg and thigh, often while lightly touching adjacent surfaces or objects. The long head of the biceps femoris muscle was silent as she stood with full extension of the hip and knee joints. However, when she stood with knee flexure, high and moderate potentials occurred in the long head of the biceps femoris muscle. As the African apes shifted additional weight onto the right hindlimb, EMG activity increased to (or remained at) moderate or high levels in the right rectus femoris, vastus lateralis, vastus intermedius, biceps femoris, semi-
G
c
2
2
1
Semitendinosus
Semimembranosus
2
G
+,+ +,+ + +
+,++
O,+,++
+
O,[+,++l
+,+ + 0
++,+++ OS+I
O,+,+
0, +
O,+,++
++
++
+,I++I
0
++.+++ ++
+,+ +
0
+ +,l++ + I
0
O,+,++
0
[+,++I
0
-
0
0, +
O.+.+
+ +
0
++
0
+,++
0
-
-
-
-
~
-
0
0
-
+++ ++ 0,+,+ +
+
0
-
0
+++
+++ ++ ++,+++
-
0
-
-
++,+++
~~
+++
-
0
+ +.+ + + -
-
-
++
Stance phase
~
-
+,+ +
+,++.++ +
0
-
++
4
-,
0
~~
0, +
0
+, low
~~
O,+,[+ +I
0
0
0
n
-
0
0
-
0
0
+,+ +
-
0
0
0
no data, 0.silence;
+,+++
0
+,+ +
+t
+,+ +
0
0
O,+,++,+++
~~
++
-
0
Swing phase
Walking
+ + , + + + +,++,+++ +,++,+++ +,++
Leap
+ +,+ + +
-
~
++
-
-
-
+,++
++
++
0, +
0
0
++
0
+,+ t
+,++,+++
o,++,+++
I+.+
++ +,++
0,+,[0,+,++1
+,++
++,+++
+.+ + +1 ++,+++ +,++,+++
+,+ +,+ + + +,++,+++
0
[0,+,++1
+,+ +
++,[+,+++I
++,+++,lo1
Lunge
+,+ + +,++ +,+ +
+,++
Digitigrade reach
Symbols: ', proxima1 one.third of thigh; t, middle one-third of thigh; G, gorilla. C, chimpanzee; 0.orangutan; N , number of 60 120minute experiments; S, subject; EMG; + C , moderate EMG; + + +, high EMG; [ I , indicate relatively rare occurrences or mainly while the subject was groggy.
Gracilis
c
G
1
1
G
It
Adductor longus
G
2*
Adductor magnus vrovi-ius
G
1
Adductor magnus (long head)
G
1
O,+,++
o,++,+++
0
1
Biceps femoris (short head)
G
+,+ t
c
2
Biceps femoris (lone head)
+,lo,+]
+
G
0
2
3
O,+,+
c
O,+,[O,+,++l
+
+,+ t
G
1
c
G
1
1
G
++,+++
+,++,+++
0
1
3
-
+
c
+,++,l+++l
+,++,+++
+,++,I01
G
Plantigrade reach
Eccentric loading stance
3
S
Quiet symmetric stance
2
N
Vastusintermedius
Vastus medialis
Vastus lateralis
Rectus femoris
Muacle
Actiuities ofpongid thigh muscles during bipedal behavior
TABLE 1
cn
til
CL
0
-
126
R. H. TU’ITLE, J. V. BASMAJIAN AND H. ISHIDA
a. b. c. d.
Fig. 1 Gorilla female. Standing on the apex of a ramp with her eccentrically loaded right hindlimb extended. Standing on a platform with her eccentrically loaded right hindlimb flexed. Standing with the left foot plantigrade and the right foot “on points” but bearing little weight. Walking bipedally with her flexed left hindlimb in mid stance phase and her right foot approaching toe off.
EMG OF PONGID THIGH MUSCLES AND BIPEDALISM
a. b. c. d.
Fig. 2 Chimpanzee male. Standing bipedally with his feet plantigrade and flexure of both hindlimbs. Walking bipedally while reaching overhead. Reaching for a food sill with his extended right hindlimb eccentrically loaded. Reaching for a suspended food incentive with his left foot gripping the edge of a ramp and with his right foot "on points."
Fig. 3 Orangutan female. a. Reaching overhead with flexure of her hindlimbs and weight resting mostly on her heels and lateral soles. b. Reaching for incentives on a sill with her hindlimba extended. c. Standing on the apex of a ramp with her eccentrically loaded right hindlimb nearly fully extended.
127
128
R. H. TUTTLE, J . V. BASMAJIAN AND H. ISHIDA
Fig. 4 EMG potentials from the right rectus femoris (a), vastus lateralis ib) and ischiofemoralis (c) muscles of a bipedal gorilla as she shifted weight eccentrically onto her flexed right hindlimb and pivoted clockwise on her right foot (secs. 1-21; during the stance phase of a bipedal step forward (secs. 2-3);during a relatively quite symmetrical stance with hip and knee flexure (secs. 4-8); during the stance phase of a slow forward step with the foot quasi-plantigrade while reaching overhead (secs. 8-9);and as she stood “on points” while manipulating a trapeze bar (secs. 10-12).(t, time in seconds. The vertical line between secs. 3 and 4 indicates that the illustration is based on discrete episodes during the recording session .)
membranosus (gorilla only), adductor magnus and adductor longus muscles. The vastus medialis muscle of the chimpanzee and the semitendinosus and gracilis muscles of the gorilla usually remained silent when the right hindlimb was eccentrically loaded during bipedal stance. In the orangutan, the rectus femoris muscle exhibited moderate and high potentials and the long head of the biceps femoris muscle was silent during eccentric loadings of the fully extended hindlimb.
Reaching The subjects used a variety of hindlimb postures when reaching overhead for the trapeze and food incentives. The hip and knee joints could be flexed (figs. lb, 2c,d, 3a,c) or extended (figs. la, 3b). The feet could be plantigrade (with heel contact: figs. la,b, 2c, 3a,c).
quasi-plantigrade (with dorsiflexion at the mid-tarsal joint so that the posterior heel was free of the substrate), or digitigrade (with the ankle markedly volarflexed and the digits extended so that only their distal pads were in contact with the floor: figs. 2d, 3b) which we have termed on points (Tuttle et al., ’75: p, 257). In table 1,plantigrade reach also includes incidents wherein the African apes used quasi-plantigrade foot postures. The hindlimbs were either somewhat flexed or fully extended during plantigrade reaches. The hips and knee joints were fully extended or overextended (i.e., beyond 180’) during high reaches with both feet on points. In the gorilla, plantigrade reaching was accompanied by low or moderate potentials in the rectus femoris, vastus lateralis, vastus intermedius, biceps femoris, semimembranosus, adductor magnus, and adductor longus mus-
129
EMG OF PONGID THIGH MUSCLES AND BIPEDALISM
C
1
2
3
4
5
6
7
8
9
sec. Fig. 5 EMG potentials from the right gracilis la), proximal adductor nlagnua proprius (bl, mid adductor magnus proprius (c). and short head of the biceps femoris !d) muscles of a bipedal gorilla. Secs. 1-2:a bipedal steps forward in which a acts during swing phase and b, c and d act during stance phase. Sec. 2-- vertical line t h e subject steps forward and stands bipedally with hip and knee flexure and hoth feet plantigrade; a acts mainly during the swing of the foot and b, c and d act while t h e hindlimb is loaded. Secs. 8-9: a leap to the trapeze from a crouched posture in which t h e thighs were widely abducted. (t and vertical line as in fig. 4.)
cles. High potentials occurred quite rarely in the reaching gorilla's rectus femoris muscle and then only when her knee was flexed. Once her semimembranosus muscle exhibited high potentials as she stumbled heavily onto the flexed right hindlimb while reaching overhead. Low and nil potentials occurred in the vastus medialis muscle. The semitendinosus and gracilis muscles were always silent during plantigrade and digitigrade reaching. As the gorilla rose t o the on points position (fig. 41,EMG levels commonly increased in the rectus femoris muscle. Sometimes while the
knee was fully extended, moderate and high potentials persisted; but at other times EMG activity decreased. The vastus medialis muscle also exhibited increased EMG activity when the gorilla reached in the digitigrade posture with full extension of her knees. There were no consistent differences in the EMG levels of the vastus lateralis, short head of the biceps femoris, semimembranosus and long head of the adductor magnus muscles. Sometimes EMG levels decreased in the vastus lateralis muscle after the gorilla had grabbed the food sill or trapeze. EMG levels often de-
1
2
3
4
5
6
7
8
9
10
11
12
sec. Fig. 6 EMG potentials from the right long head of the adductor magnus (a), gracilis (b), and adductor longus (c) muscles of a bipedal gorilla. Secs. 1-6and 10-12: a and c act during the stance phase of bipedal steps as the subject walks about while reaching overhead. Secs. 7-9:the subject lunges for the trapeze uligh potentials just after sec. 7) and then stands bipedally with her thighs widely abducted.
creased dramatically in the adductor magnus proprius muscle, especially in its middle onethird, as the gorilla achieved full extension of the knee during digitigrade reaching. In the chimpanzee, both plantigrade and digitigrade reaching were accompanied by low and moderate potentials in the rectus femoris and vastus lateralis muscles. Most reaching occurred beneath the food sill where the chimpanzee frequently hoisted himself bimanuall y or by the right forelimb alone instead of remaining bipedal on the floor. Thus, some of the observed decreases in EMG levels in the rectus femoris and vastus lateralis muscles as he achieved full extension of the hindlimbs may have occurred because he had begun to hoist himself, thereby lessening the load on his feet. Low, high, or most commonly moderate potentials occurred in the long head of the biceps femoris muscle during plantigrade reaches. Since (when not hoisting) he preferred to reach for the food sill with his left hand, often his right hindlimb was eccentrically loaded (fig. 2c). The chimpanzee's semimembranosus muscle was silent during all reaches. His adductor magnus proprius muscle was usually
silent during reaching though occasionally low or moderate potentials occurred during plantigrade reaching. In the orangutan, plantigrade reaching with the hip and knee joints fully extended was accompanied by moderate and high potentials in the rectus femoris muscle and silence in the long head of the biceps femoris muscle. The latter muscle was also silent during digitigrade reaching in which the hindlimb was fully extended.
Lunging and leaping Lunges (wherein the distal portions of the feet remained on the floor) and leaps (wherein the feet rose completely off the floor) were executed by the African apes as they attempted to grab the trapeze and food incentives on a sill or wall. The chimpanzee lunged and leaped much more frequently than the gorilla did. In the gorilla, moderate potentials occurred in the rectus femoris and short head of the biceps femoris muscles during lunges and moderate and high potentials accompanied leaps (fig. 5). The semitendinosus and gracilis
EMG OF PONGID THIGH MUSCLES AND BIPEDALISM
131
C
1
2
3
4
5
6
8
1
see. Fig. 7 EMG potentials from the right gluteus medius (a), rectus femoris (b), and adductor magnus proprius (c) muscles of a bipedal chimpanzee. The subject rises from a quadrupedal to a bipedal posture (sec. 1) and takes two steps (sees. 1-31. a and b act during the stance phase of each step. c acts as the subject rises (aec. 1) and then during the swing phase of each step (prior to secs. 2 and 3). Sec. 3- vertical line: subject stands with his thighs widely abducted and with his right knuckled hand propped on a wall. Secs. 7-8: vigorous leap from a crouched knuckle-walking posture to a trapeze bar. Note that potentials begin in a (associated with erecting the trunk by hip extension) before b commences activity (associated with propulsive knee extension). (t and vertical line as in fig.4.)
muscles were silent during lunges and the gracilis muscle was silent during leaps. The gorilla only lunged once while we were recording the long head of the adductor magnus muscle. It produced low potentials. The proximal one-third of the adductor magnus proprius muscle exhibited high potentials and its middle one-third exhibited moderate potentials during all lunges and leaps (fig. 5). Leaps and lunges were accompanied by high potentials in the adductor longus muscle (fig. 6 ) . In the chimpanzee, the rectus femoris muscle exhibited potentials ranging from low to moderate during lunges. Leaps were accompanied commonly by high or moderate potentials and rarely by low potentials in the rectus femoris muscle (fig. 7). The higher potentials were especially characteristic of leaps for the trapeze where, unlike leaping below the food sill, he could not augment the propulsive ef-
fort of the hindlimbs by hoisting. Lunges were accompanied by moderate potentials in the vastus lateralis muscle. Moderate potentials also occurred during most leaps; but high potentials were exhibited by the vastus lateralis muscle during especially vigorous leaps for the trapeze. Per contra the vastus medialis was silent during leaps. The semimembranosus muscle was also typically silent during leaping and lunging. When the chimpanzee lunged for the trapeze or food sill, the long head of the biceps femoris muscle exhibited moderate or high potentials. All leaps for the trapeze were accompanied by high potentials in the long head of the biceps femoris muscle. The adductor magnus proprius muscle exhibited quite variable EMG levels, ranging from nil to moderate potentials, during lunges. Leaps were characteristically accompanied by moderate or high potentials in the
132
R. H. TUTTLE, J. V. BASMAJIAN AND H. ISHIDA
chimpanzee's adductor magnus proprius muscle (fig. 7 ) . Bipedal walking The two African apes exhibited similar patterns of bipedal walking. They did not differ remarkably from descriptions of other bipedal chimpanzees and gorillas (Elftman, '44; Tuttle, '70; Jenkins, '72; Ishida et al., '75). Throughout each locomotive cycle, their knee and hip joints were flexed and the knee did not pass posteriorly beyond the ipsilateral hip joint. There was no evidence of pelvic tilt of a hominid sort. The subjects alternately leaned laterally over each supportive foot producing a waddling gait (Tuttle and Basmajian, '75; Tuttle et al., '75, '78). Many thigh muscles were quite active during the stance phase of the bipedal locomotor cycle in the African apes. In all instances, the activity consisted of uniphasic bursts of potentials which occurred predominantly during load bearing and propulsive segments of the stance phase. In the gorilla, moderate or high potentials usually or always occurred in the rectus femoris, vastus lateralis, vastus intermedius, proximal one-third of the adductor magnus proprius, and the adductor longus muscles during the stance phases of bipedal steps (figs. 4-6). Low potentials occasionally occurred in the rectus femoris and vastus intermedius muscles during some steps, especially when the gorilla touched a wall or other supportive objects. The short head of the biceps femoris muscle exhibited potentials ranging from nil t o high, with low and moderate ones being most common, during the stance phase of bipedal steps (fig. 5 ) . Once as the gorilla took several steps backwards, only inconsequential single potentials occurred in the short head of the biceps femoris muscle during the stance phase of each step. The long head of the adductor magnus and the middle one-third of the adductor magnus proprius muscles usually exhibited low and occasionally moderate potentials during the stance phase of bipedal steps (figs. 5, 6). The gorilla's semitendinosus and gracilis muscles were silent during the stance phase of bipedal steps (figs. 5 , 6). All thigh muscles which we recorded in the gorilla, except the gracilis, were silent during the swing phase of bipedal locomotor cycles. The gracilis muscle exhibited low and inconsequential single potentials during the swing phase of some steps (figs. 5,6).
In the chimpanzee, the rectus femoris and vastus lateralis muscles usually exhibited moderate and, in some instances, low potentials during the stance phase of bipedal steps (fig. 7). The long head of the biceps femoris muscle also exhibited low and moderate potentials during the stance phase of bipedal steps. Generally EMG activity was lower in the biceps femoris muscle during bipedal walking than during lunging, leaping, reaching and standing with the right hindlimb eccentrically loaded. The semimembranosus and adductor magnus proprius muscles were silent during the stance phase of all bipedal locomotor cycles (fig. 7). Two of the six thigh muscles recorded in the chimpanzee exhibited EMG activity during the swing phase of bipedal steps. Brief bursts of moderate and low potentials occurred in the vastus medialis muscle during swing phase, just prior to foot contact. The vastus medialis muscle was silent during all of stance phase. Thus it acted quite discretely from the vastus lateralis and rectus femoris muscles. Although the swing phases of most steps were accompanied by silence in the adductor magnus proprius muscle, a few were accompanied by low potentials and moderate potentials occurred during one step (fig. 7 ) . DISCUSSION
Comparisons with other EMGStudies on non-human primates Using surface electrodes, Ishida et al. ('75) recorded EMG activity in the vastus lateralis and long head of the biceps femoris muscles in a 7.5- t o 8-year-old female chimpanzee which had been trained over a 6-year period to walk bipedally on a level platform. Both muscles acted during the stance phase of bipedal steps. This is basically coincident with our results on a younger male chimpanzee, which had not experienced an extended period of bipedal training. Further, the actions of the vastus lateralis muscle in Pan gorilla and Pan troglodytes and of the long head of the biceps femoris muscle in P. troglodytes are not notably different from actions of their counterparts in Hylobates agilis, Macaca fuscata, Papio harnadryas, and Ateles geoffroyi during bipedal walking (Ishida et al., '75). In all species, they are active during the stance phase of each step (fig. 5 in Ishida et al., '75). Because the methods of recording and analyzing data differ between our study and that of Ishida et al. ('751, we cannot ascribe major functional impor-
EMG OF PONGID THIGH MUSCLES AND BIPEDALISM
tance to apparent dissimilarities in levels of EMG activity between them, With indwelling fine-wire bipolar electrodes, Ishida et al. ('78) recorded the activities of many thigh and hip muscles in a juvenile male HyEobates agiEis and a juvenile female Hylobates lar as they walked bipedally on a level platform. The hamstring, quadriceps femoris, sartorius, gracilis, adductor longus, adductor magnus, and gluteal muscles were quite active and the iliopsoas muscle exhibited minor activity during the stance phase of bipedal steps. The iliopsoas and rectus femoris muscles of Hylobates were active during most of (especially mid) swing phase and the gracilis and adductor longus muscles were active during release of the foot and early swing phase. Ishida et al. ('75) concluded that the latter two muscles initiate the swing phase by slightly raising the thigh. Then the iliopsoas and rectus femoris muscles carry the limb forward. The rectus femoris muscle also extends the knee joint to reposition the foot during the swing phase. The vastus lateralis muscles of both African apes, the vastus intermedius and adductor magnus muscles of the gorilla, and the long head of the biceps femoris muscle of the chimpanzee act basically like their counterparts in gibbons during bipedal progression. As in gibbons, the rectus femoris muscles of the African apes and the adductor longus muscle of the gorilla are active during the stance phase. But they differ from their counterparts in the gibbons in that they are silent during the swing phase of bipedal steps. By remaining silent during the stance phase of bipedal steps, the vastus medialis, semimembranosus and adductor magnus proprius muscles of the chimpanzee and the semitendinosus and gracilis muscles of the gorilla differ strikingly from their counterparts in the gibbons. This might be attributed t o the fact that the gibbons usually moved rapidly whereas the African apes progressed more slowly. It is also possible that the African apes walk in a manner which loads the medial thigh relatively less than in the gibbons. The gorilla resembles the gibbons somewhat in actions of the gracilis muscle during the swing phase of bipedal steps. The activity of the vastus medialis muscle during the late swing phase in the chimpanzee contrasts his bipedal walking with the bipedalism of the gorilla and two gibbons. EMG patterns from the gluteal muscles (gluteus maximus proprius, ischiofemoralis,
133
gluteus medius and gluteus minimus) of the two gibbons are basically similar to those of the gorilla and chimpanzee, viz. notable activity during the stance phase and silence during the swing phase of bipedal steps (Ishida et al., '78; Tuttle and Basmajian, '75; Tuttle et al., '75, '78). Functional implications Our studies of EMG activity in the thigh and gluteal muscles in African apes during bipedal stance partly confirm Kummer's ('75: pp. 285-2861 prediction t h a t considerable gluteal and hamstring activity would be required in order for a chimpanzee to stand bipedally with flexure of the hips and knees. The gluteus maximus proprius, ischiofemoralis, gluteus medius (fig. 7) and gluteus minimus muscles generally exhibit low or nil potentials during quiescent bipedal stance, though sometimes higher potentials occur, especially in the ischiofemoralis and gluteus minimus muscles (Tuttle et al., '78). This modest activity of the gluteal muscles to maintain hip extension (figs. 4,7) is especially complemented by low and moderate activity in the biceps femoris muscle (on the lateral aspect of the thigh) in both African apes and by the semimembranosus muscle (on the medial aspect of the thigh) in the gorilla. The other thigh muscles which exhibit notable potentials during bipedal stance probably act mainly t o sustain the partially flexed knee joint and the laterally rotated and abducted thigh against gravitational forces. As bipedal African apes fully extend their knee and hip joints, thereby reducing the moment of the extensor muscles (Kummer, '75: p. 2861, some muscles act quite prominently, especially when the on points posture is employed. Of the muscles which we recorded, this is particularly true for the gluteus maximus proprius (Tuttle et al., '75, '781, rectus femoris, vastus lateralis, vastus intermedius, and perhaps the biceps femoris. Not surprisingly, in the African apes many hip and knee extensor muscles are very active during the propulsive thrust of lunges and leaps. Often the subjects leap from crouched postures in which the thighs are flexed, abducted and laterally rotated and the knees are markedly flexed. Of the muscles which we recorded in the chimpanzee and gorilla during leaps and lunges, the gluteus maximus proprius, ischiofemoralis, gluteus medius (fig. 7), gluteus minimum (Tuttle et al., '781, and
134
R.H. TU'ITLE, J . V . BASMAJIAN AND H. ISHIDA
biceps femoris (long head) are major hip extensors; the adductor magnus and adductor longus are major adductors of the hip joint; and the rectus femoris and vastus lateralis are major knee extensors. Elftman ('44: p. 70) suggested that during the mid stance phase of bipedal walking in the chimpanzee the hip extensors must be active in order to restore the elevation of the body while concurrently increasing its forward velocity. This prediction is confirmed by our observations of notable activity in the gluteus maximus proprius, ischiofemoralis (fig. 4), gluteus medius (fig. 71, gluteus minimus (Tuttle et al., '781, and biceps femoris (long head) muscles during the stance phases of most bipedal steps by t h e African apes. In the gorilla, the long head of the adductor magnus and perhaps the adductor magnus propius and adductor longus muscles might also contribute to hip extension during the stance phase of bipedal steps. Comparisons with m a n
The EMG patterns exhibited by thigh and hip muscles in man and apes are generally strikingly dissimilar during most bipedal behavior and especially during bipedal progression. Human quiescent bipedal stance is remarkably economical in that no thigh and probably only a few hip muscles are active (Basmajian, '74; Carlsoo, '72; Okada, '72, '75). Integrity of the extended hip and knee joints is largely premised on osseoligamentous mechanisms in plantigrade standing humans. Further, Okada ('72, '75) reported that even when human subjects rise onto the "balls" of their feet, the gluteus maximus, rectus femoris, vastus lateralis, vastus medialis, semitendinosus and biceps femoris muscles remain silent or evince very low EMG activity. Per contra, quite high EMG activity occurs regularly in the rectus femoris muscle of the orangutan when i t stands bipedally with full alignment of the hindlimb. In plantigrade human subjects, flexure of the knee and hip joints to positions approximating those of bipedal pongid apes is accompanied by major increases in the activity of the quadriceps femoris and biceps femoris muscles, modest increases in the activity of the gluteus maximus muscle and little change in the semitendinosus muscle (Okada, '72, '75). Like human stance and unlike pongid bipedalism, human bipedal walking is accom-
panied by remarkably little EMG activity in the thigh muscles (Basmajian, '74: p. 308). During bipedal walking, human subjects also differ strikingly from pongid apes and other anthropoid primates in the much greater frequency of biphasic and triphasic EMG patterns evinced by human thigh muscles (Basmajian, '74: pp. 323-325; Carlsoo, '72: p. 117) versus uniphasic ones in the non-human subjects. In human subjects, the quadriceps femoris and hamstring muscles act from mid to late swing phase through heel strike and early stance (initial double support) phase and again during mid stance (single support) phase. The human adductor muscles act during late swing phase through initial double support phase and again from second double support phase into early swing phase (Carlsoo, '72: p. 117). The evolutionary problem of hominid bipedalism: what d o we need i n order toproceed ? Commentators on the evolution of hominid bipedalism often emphasize (and indeed marvel a t ) the achievement of balance and the economy of muscular activity in the human lower limb. Our findings are in general agreement with this perspective. Yet because human kinesiological studies are focussed on sexually mature subjects we are ignorant about how and when this condition is developed ontogentically. Additional studies on ad libitum human running, sustained terrestrial bipedal locomotion in apes and arboreal walking and running by apes are also needed before we can begin to sketch the many small steps in hypothetical evolutionary sequences which might have culminated in bipedal Homo sapiens. Experimental studies on arboreal bipedalism and vertical climbing in apes might prove to be particularly informative re possible preadaptations for hominid terrestrial bipedalism. The neurological bases for complex biphasic and triphasic activities of human lower limb muscles should also be explored in detail. Is there central pattern generation related to bipedalism in Homo sapiens? To what extent is this homologous with the neurological substrates for bipedal or other positional behavior in non-human primates and especially with neuromuscular mechanisms in man's closest living relatives, the apes? Finally, we should mention that whatever hypotheses we might formulate on the basis of
EMG OF PONGID THIGH MUSCLES AND BIPEDALISM
kinesiological studies of living primates, we must have the skeletons of ante- and early hominids in order to test their validity. It now seems more likely than ever t h a t such materials will be available before or a t the time that the challenging kinesiological studies mentioned herein above are implemented. ACKNOWLEDGMENTS
This study was supported mainly by NSF Grants GS-3209 and SOC75-02478 and by a Public Health Service Research Career Development Award (1-K04-GM16347-01) from the National Institutes of Health. Supplementary support was provided by NIH Grant RR00165 to the Yerkes Regional Primate Research Center and the Marian and Adolph Lichtstern Fund of t h e University of Chicago. We are especially thankful for the assistance of Doctor G. H. Bourne, J. Malone, E. Regenos, J. Perry, R. Pollard, S. Lee, R. Mathis, J. Roberts, Doctor M. Keeling, Doctor M. Vitti, J. Hudson, K. Barnes and J. Rike. This paper is dedicated to the memory of Mrs. Almas Benjamin, who died on March 12, 1978. She loved primates and endeavored to impart an appreciation of them to pupils a t the Christopher House Day Care Center in Chicago, Illinois. LITERATUHE CITED Basmajian, J. V. 1974 Muscles Alive. Their Functions Revealed by Electromyography. Third ed. The Williams & Wilkins Co., Baltimore. CarlsM, S. 1972 How Man Moves. Kinesiological Methods and Studies. Wm. Heinemann Ltd., London. Elftman, H. 1944 The bipedal walking of t h e chimpanzee. J. Mamm., 25: 67-71. Ishida, H., T. Kimura and M. Okada 1975 Patterns of bipedal walking in anthropoid primates. In: Symp. 5th Cong. Int’l. Primat. SOC.(1974). S. Kondo, M. Kawai, A. Ehara and S. Kawamura, eds. Japan Science Press, Tokyo, pp. 287-301. Ishida, H., M. Okada, R. H. Tuttle and T. Kimura 1978 Ac-
135
tivities of hindlimb muscles in bipedal gibbons. In: Recent Advances in Primatology. Vol. 3. D. 3. Chivers and K. A. Joysey, eds. Academic Press, London, pp. 459-462. Jenkins, F. A,, Jr. 1972 Chimpanzee bipedalisrn: cineradiographic analysis and implications for the evolution of gait. Science, 178: 871-879. King, M.-C., and A. C. Wilson 1975 Evolution a t two levels in humans and chimpanzees. Science, 188: 107-116. Kummer, B. K. F. 1975 Functional adaptation to posture in t h e pelvis of man and other primates. In: Primate Functional Morphology and Evolution. R. H. Tuttle, ed. Mouton, The Hague, pp. 281-290. Miller, D. A. 1977 Evolution of primate chromosomes. Science, 198: 1116-1124. Okada, M. 1972 An electromyographic estimation of the relative muscular load in different human postures. J. Human Ergo]., 1: 75-93. 1975 Quantitative studies on t h e bearing of t h e anti-gravity muscles in human postures with special references to electromyographic estimation of t h e postural muscle load. J. Fac. Sci., Univ. Tokyo, Sec. V, Vol. IV, pp. 471-530. Sarich, V. M., and J. E. Cronin 1976 Molecular systematics of the primates. In: Molecular Anthropology. M. Goodman and R. E. Tashian, eds. Plenum, New York, pp. 141-170. Tuttle, R. H. 1970 Postural, propulsive and prehensile capabilities in the cheiridia of chimpanzees and other great apes. In: The Chimpanzee, Vol. 2. G. H. Bourne, ed. Karger, Basel, pp. 167-253. Tuttle, R. H., and J. V. Basmajian 1974a Electromyography of brachial muscles in Pun gorilla and hominoid evolution. Am. J. Phys. Anthrop., 41: 71-90. 1974b Electromyography of forearm musculature in gorilla and problems related to knuckle-walking. In: Primate Locomotion. F. A. Jenkins, Jr.,ed. Academic Press, New York, pp. 293-347. 1975 Electromyography of Pan gorilla: a n experimental approach to t h e problem of hominization. In: Symp 5th Cong. Int’l. Primat. Soc. (1974). S. Kondo, M. Kawai, A Ehara and S. Kawamura, eds. Japan Science Press, Tokyo, pp. 303-314. Tuttle, R. H., J. V. Basmajian and H. Ishida 1975 Electromyography of the glUteUR maximus muscle in gorilla and t h e evolution of bipedalism. In: Primate Functional Morphology and Evolution. R. H . Tuttle, ed. Mouton, The Hague, pp. 253-269. 1978 Electromyography of pongid gluteal muscles and hominid evolution. In: Recent Advances in Primatology. Vol. 3. D. J. Chivers and K. A. Joysey, eds. Academic Press, London, pp. 463-468.
