Hind limb extensor muscle architecture reflects locomotor specialisations of a jumping and a striding quadrupedal caviomorph rodent

Abstract

Muscle architecture is an important factor in determining muscle function. The physiological cross-sectional area (PCSA) is directly proportional to the force-generating capacity of a muscle, while fibre length determines the capacity for a muscle’s length change. For a given muscle volume, both parameters cannot be maximised at the same time, and therefore, specialisation in accordance with specific functional demands is widely accepted. Building on this, the architecture of selected hind limb extensor muscles of two caviomorph rodent species of similar body size but with differing locomotor modes were analysed and compared. Individual fascicles of fixed cadavers were carefully removed during stepwise dissection. After removal of each fascicle, the left-behind groove within the muscle belly was digitised to capture the length and orientation of the removed fascicle. Pennation angle, muscle volume, and anatomical cross-sectional area were determined, and finally, PCSA and force-generating capacity were approximated for a hip extensor (M. biceps femoris), a knee extensor (M. vastus lateralis), and an ankle extensor (M. triceps surae). Muscle architecture appeared to reflect locomotor specialisation of a jumping (Chinchilla chinchilla) in comparison with a striding quadrupedal (Cavia porcellus) species, but considerable variability of the limited specimens analysed was found. With the biceps femoris as an exception, analysed specimens of Chinchilla had relatively more voluminous and thus metabolically expensive hind limb extensors with both a greater capacity for length change and for force generation. These results are in agreement with a greater demand for powerful hind limb extension during launches and provide further evidence that muscle architecture is adapted to differing functional demands in closely related species.

Full text

Vol.:(0123456789)
1 3
Zoomorphology
DOI 10.1007/s00435-017-0349-8
ORIGINAL PAPER
Hind limb extensor muscle architecture reflects locomotor
specialisations ofajumping andastriding quadrupedal
caviomorph rodent
SusannRosin1· JohnA.Nyakatura1 
Received: 21 December 2016 / Revised: 13 February 2017 / Accepted: 16 February 2017
© Springer-Verlag Berlin Heidelberg 2017
and for force generation. These results are in agreement
with a greater demand for powerful hind limb extension
during launches and provide further evidence that muscle
architecture is adapted to differing functional demands in
closely related species.
Keywords Fascicle· Functional morphology· Hind
limb· Mammal· Caviomorpha· Guinea pig· Chinchilla
Introduction
Mammalian skeletal muscle function is a complex, multi-
levelled phenomenon. For example, on a macroscopic
scale, muscle topography has direct influence on the poten-
tial torque generated at joints and on potential out forces
generated at end-effectors, because the position of muscle
insertions on the bones determines a muscle’s moment arm
to a joint’s centre of rotation. In contrast, on a microscopic
scale, differential staining revealed that mammalian skeletal
muscle is composed, to varying proportions between motor
units, of fibres of either slow-oxidative, fast glycolytic or
intermediate fibre-type (e.g., Biewener 2003). A single
motor unit consists of a motor neuron, the axonal terminals,
and the sum of all fibres that are activated by this motor
neuron. A muscle usually consists of several motor units
and not all of them are typically activated at the same time.
Due to a differential activation of motor units that are made
up of fibres with differing metabolic and contractile proper-
ties within a single muscle, a muscle can be functionally
compartmentalised without this being visible on the macro-
scopic level (Scholle etal. 2001). On an intermediate scale,
muscle architecture, i.e., the geometric properties and the
arrangement of the contractile units within a muscle, has
fundamental significance for the functioning of the muscle,
Abstract Muscle architecture is an important factor in
determining muscle function. The physiological cross-sec-
tional area (PCSA) is directly proportional to the force-gen-
erating capacity of a muscle, while fibre length determines
the capacity for a muscle’s length change. For a given
muscle volume, both parameters cannot be maximised at
the same time, and therefore, specialisation in accordance
with specific functional demands is widely accepted. Build-
ing on this, the architecture of selected hind limb extensor
muscles of two caviomorph rodent species of similar body
size but with differing locomotor modes were analysed and
compared. Individual fascicles of fixed cadavers were care-
fully removed during stepwise dissection. After removal
of each fascicle, the left-behind groove within the muscle
belly was digitised to capture the length and orientation
of the removed fascicle. Pennation angle, muscle volume,
and anatomical cross-sectional area were determined, and
finally, PCSA and force-generating capacity were approxi-
mated for a hip extensor (M. biceps femoris), a knee exten-
sor (M. vastus lateralis), and an ankle extensor (M. triceps
surae). Muscle architecture appeared to reflect locomotor
specialisation of a jumping (Chinchilla chinchilla) in com-
parison with a striding quadrupedal (Cavia porcellus) spe-
cies, but considerable variability of the limited specimens
analysed was found. With the biceps femoris as an excep-
tion, analysed specimens of Chinchilla had relatively more
voluminous and thus metabolically expensive hind limb
extensors with both a greater capacity for length change
* John A. Nyakatura
john.nyakatura@hu-berlin.de
1 AG Morphologie und Formengeschichte, Institut für Biologie
& Bild Wissen Gestaltung. Ein interdisziplinäres Labor,
Philippstraße 12, 10115Berlin, Germany

Zoomorphology
1 3
too (e.g., Gans and Bock 1964; Spector etal. 1980; Sacks
and Roy 1982; Wickiewicz et al. 1983, 1984; Fukunaga
etal. 1997; Lieber and Fridén 2001; Payne etal. 2005).
Especially two parameters—the fascicle length and the
physiological cross-sectional area (PCSA)—are of para-
mount functional significance on the scale of muscle archi-
tecture. The fascicle length reflects the ‘working range’
of the muscle, whereas the PCSA is directly proportional
to the ‘muscle force’ (cf. Lieber and Fridén 2001; Allen
etal. 2010, 2014; Dick and Clemente 2016). Muscle length
change is a function of how many contractile units (i.e., the
sarcomeres within the fibres that make up the fascicles) are
arranged in series. In addition, the number of sarcomeres
arranged in series predominantly determines the velocity of
contraction (Lieber and Fridén 2001). On the other hand,
force producing capacity is a function of how many sar-
comeres are arranged in parallel (Wickiewicz etal. 1983).
Together, fascicle length and PCSA largely determine how
much work (force × distance) can be done by a muscle and
how much power (work/time) can be produced (cf. Payne
etal. 2005; Allen etal. 2010, 2014). For a given volume,
both parameters cannot be maximised at the same time
(Wickiewicz etal. 1983; Epstein and Herzog 1998; Allen
etal. 2010). A highly powerful muscle has long fascicles,
a large PCSA, and hence a large volume. However, volumi-
nous, powerful muscles also are metabolically expensive,
and therefore, it can be expected that mammalian skeletal
muscles are as small as possible without impairing the
overall function and biological role within the organism.
Consequently, for a muscle of a given volume, a trade-off
can be expected between a specialisation for force-gener-
ating capacity (relatively large PCSA) and a specialisation
for working range (relatively long fascicles).
A third important muscle architectural parameter is
the pennation angle (θ) of the contractile fibres relative to
the direction of force generation of the entire muscle. A
larger pennation angle allows for the packing of more mus-
cle fibres within a given volume which results in a higher
PCSA and hence a relatively larger force-generating capac-
ity (e.g., Allen et al. 2010). At the same time, this larger
capacity for force generation is partly offset, because in
theory, any θ > 0° will result in a loss of force along the line
of action when compared to a muscle with the same mass
and fibre length but with zero pennation angle (Lieber and
Fridén 2001). A larger θ also consequently results in rela-
tively shorter fascicles per volume, resulting in a relatively
reduced capacity for overall muscle length change.
Taking into account these now well accepted mechani-
cal relationships, it can be expected that muscle architec-
ture reflects adaptations of the musculoskeletal system in
accordance with differing functional demands in closely
related species. In fact, architectural specialisations have
been demonstrated for muscles with different functional
demands within one organism (e.g., Payne et al. 2006;
Moore etal. 2013; Rupert etal. 2015) and also in accord-
ance with differing functional demands posed by changes
in body mass during ontogeny (e.g., Allen et al. 2010)
and phylogeny (e.g., Allen etal. 2014; Dick and Clemente
2016). To further investigate this notion, we here stud-
ied the muscle architecture of selected hind limb exten-
sor muscles of two closely related South American mam-
mal species of similar body size, but differing locomotor
regimes—the guinea pig (Cavia porcellus Lichtenstein
1829, Caviomorpha, Rodentia) and the chinchilla (Chin-
chilla chinchilla Lichtenstein 1829, Caviomorpha, Roden-
tia). Cavia is a quadruped using symmetrical gaits (Rocha-
Barbosa etal. 2005), whereas Chinchilla generally uses the
bounding gait with simultaneous hind limb powered jumps
(Spotorno etal. 2004; Elissamburu and Vizcaino 2004). C.
porcellus likely is the domesticated form of Cavia tschu-
dii and does not exist in the wild (Spotorno etal. 2007). C.
tschudii is a crepuscular social species common in South
American grasslands (Nowak 1999). In contrast, C. chin-
chilla is an endangered nocturnal species that lives in rocky
and arid high-altitude habitats in the Andes Mountains
(Roach and Kennerley 2016). For this study, we used the
domesticated form of C. chinchilla. Considering the above
mentioned functional consequences of muscle architectural
properties, we expected to find the following differences
in hind limb extensor muscles in accordance with the two
diverging locomotor specialisations of a striding quadruped
and a jumper:
1. As jumping with pronounced hind limb extension
necessitates large forces (e.g., Demes etal. 1996) and
larger excursion angles during launches than quadru-
pedal striding locomotion (e.g., Essner Jr. 2002; Demes
etal. 2005; Legreneur etal. 2010), hind limb extensor
muscles are more powerful relative to body mass in
Chinchilla when compared to Cavia.
2. Hind limb extensor muscles of Chinchilla, further-
more, reflect a specialisation towards greater force gen-
eration relative to Cavia by having (a) relatively larger
PCSA and (b) relatively larger pennation angles.
3. Hind limb extensor muscles of Chinchilla finally
reflect a specialisation towards larger excursion angles
of hind limb elements during quadrupedal locomotion
by having relatively longer fascicles.
We dissected cadavers of both species. During dissec-
tions, we stepwise removed all individual fascicles (i.e.,
bundles of fibres) that make up a hip extensor (M. biceps
femoris; BF), a knee extensor (M. vastus lateralis; VL),
and an ankle extensor (M. triceps surae; TL). As powerful
extension is a key kinematic component of jumping, these
muscles were deemed to most likely reflect functionally

Zoomorphology
1 3
significant differences in muscle architecture between the
striding quadruped and the jumper. The length and orienta-
tion of all fascicles that make up a muscle were digitised
to account for the intramuscular heterogeneity of muscle
architecture and to derive the muscle architectural parame-
ters in subsequent analyses (cf. Stark etal. 2013; Nyakatura
and Stark 2015). All data were corrected for body mass
to account for the slight differences in size between all
specimens.
Materials andmethods
Two adult, male cadavers of each species were analysed
(Table1). We received the frozen specimens from a local
breeder and a local pet shop. Exact age of the animals was
unknown, but none of the specimens showed any signs of
senility or any other peculiarities. Pet animals from breed-
ers may not have had the opportunity to develop and dis-
play the whole locomotor repertoire that is characteristic of
the species. Therefore, we cannot rule out that the results
of this study may be affected by this. We expect, however,
that differences between the wild counterparts of the spe-
cies analysed here will be more pronounced for this rea-
son. After thawing, the cadavers were eviscerated, skinned,
weighed, and the cadavers were cut at the thoracic to lum-
bar transition. We used wire to mount the hindquarters on a
metal bar. For this, we chose to approximate the limb joint
angles of Cavia at mid-stance (cf. Fischer etal. 2002) with
the help of a protractor to ensure a comparable posture in
all four individuals. The bar with the wire fixed hindquar-
ters was subsequently put into 4% formalin solution (Rothi-
Histofix, Carl Roth, Karlsruhe, Germany). Afterwards, the
metal bar was mounted to a wooden board and not removed
until the digitisation of fascicles of one body side was
completed.
Three-dimensional (3D) reconstruction of the targeted
muscles (BF, VL, and TL) of both body sides was achieved
by careful dissection of individual fascicles and subsequent
digitisation of the length and orientation of the visible
groove left behind after removal of the fascicle (cf. Dumas
etal. 1988; Poelstra etal. 2000; Kim etal. 2007; Rosatelli
et al. 2008; Stark et al. 2013; Nyakatura and Stark 2015;
Siebert et al. 2015). To remove individual fascicles, we
used a small forceps and made an effort to remove fascicles
in one piece. Dissections were made while looking through
a magnifier with a mounted ring light. Prior to digitisation
of three hind limb extensor muscles, superficial fascia were
carefully removed. The M. gluteus superficialis was com-
pletely removed to reach the deeper BF. We used a Micro-
scribe M digitiser (Hadcam, Munich, Germany) equipped
with a pointed needle tip for digitisation. For calibration of
the digitised muscle architecture, metal pins were placed at
the four corners of the wooden board and on three points of
the hindquarters that were not interfering dissections. These
calibration points were digitised before and after each dis-
section session. For the data acquisition, the locations (x, y,
z coordinates) of several points (usually 5–7) within the vis-
ible left-behind groove of a fascicle were recorded to cap-
ture the length and orientation of the previously removed
fascicle. We aspired to digitise all fascicles that make up
a muscle. The raw digitised data were imported into freely
available custom software “cloud2” programmed by Heiko
Stark (http://starkrats.de). This software allows for the con-
nection of the digitised points of individual fascicles to
form lines that represent each fascicle’s length and orienta-
tion relative to the coordinates of the calibration points.
Data analysis was also conducted in cloud2. Fascicle
lengths (L) and orientation were determined from the lines
that represent fascicles. Subsequently, muscle volume (V)
and, following Epstein and Herzog (1998), the anatomi-
cal cross-sectional area (ACSA = V/L) were calculated (for
details of this procedure using the raw data from digitisa-
tion see Stark etal. 2013). However, we here accounted for
the mean pennation angle θ relative to the line of action of
the muscle (i.e., the approximated axis of force generation;
Lieber and Fridén 2001). To determine the pennation of
the fascicles that make up a muscle it was rotated within
cloud2 until the line of action of the considered muscle
was identical with the x-axis of a Cartesian coordinate sys-
tem. The y-axis was set to point medially and the z-axis
was set to point dorsally. We provide the fascicles’ penna-
tion angles relative to the x, z (sagittal) plane and the x, y
(horizontal) plane. The mean pennation angle relative to
both planes (θ) of a specimen’s analysed muscles was used
to derive the relevant PCSA (PCSA = cos θ × ACSA; also
see Powell etal. 1984). Finally, force-generating capacity
(F = k × PCSA) of a muscle was estimated (e.g., see Epstein
and Herzog 1998). The constant k denotes a muscle’s esti-
mated maximum isometric stress (Wells 1965; Zajac 1989;
Allen et al. 2010). Usually, values between 20 and 40 N/
cm² are used (Close 1972; Epstein and Herzog 1998). We
used a value of k = 22.5N/cm² in this study, which has been
proposed to serve as a nominal value of specific tension for
mammalian muscle (cf. Lieber and Fridén 2001).
Table 1 Analysed specimens
*Body mass after evisceration
**CRL crown rump length
Species Abbreviation Mass (g)* CRL (cm)**
Cavia porcellus CP1 277 22.5
Cavia porcellus CP2 233 20.0
Chinchilla chinchilla CC1 407 20.0
Chinchilla chinchilla CC2 353 21.5

Zoomorphology
1 3

Zoomorphology
1 3
While the stepwise dissection of all individual fascicles
that make up a muscle allowed to report the heterogene-
ity within muscles, it limited the overall number of mus-
cles that were analysed within the scope of this study. Data
from different individuals were not pooled. For parameters
derived for an individual’s muscle, we provide mean ± 1
standard deviation. Limited sample size prevented statisti-
cal comparisons between species, but we discuss trends in
our data qualitatively. To account for the differences in size
in our graphs, we corrected length data (fascicle length/
body mass0.33) and area data (PCSA/body mass0.66) to body
mass (Allen etal. 2010).
Results
Overall, more than 13,500 fascicles (N = 6981 for Cavia,
N = 6782 for Chinchilla) that make up the BF, VL, and TS
in the analysed specimens of both species were digitised for
this study (Fig.1). All quantified parameters are reported in
Table2 (absolute measurements).
M. biceps femoris (BF) BF is a superficial muscle that
has two distinct heads proximally, the caput longum and
the caput breve (Fig.1). The caput longum originates on
the vertebral column caudal to the hip joint. The caput
breve originates caudal to the caput longum on the pelvic
bone (ischium). In Cavia and Chinchilla fascicles of both
heads merged close to their respective origins and a sin-
gle and relatively flat muscle belly stretched over the lat-
eral aspect of the hind limb. Because fascicles within this
muscle belly did not span the whole distance between ori-
gin and insertion, it was decided against a somewhat arbi-
trary distinction between the two heads in the quantitative
analysis (below). The muscle has a broad insertion from
the lateral epicondyle of the femur to an aponeurosis along
the tibia. In Cavia the caput longum formed a large angle
to the approximated axis of force generation. In addition,
in Cavia, the insertion along the tibia reached far distally
almost to the ankle joint. In contrast, in Chinchilla, both
heads were more or less aligned with the force-generating
axis and the insertion did not reach as far distally. In both
species, a considerable share of fascicles spanned only the
hip joint and their action would have extended it (or would
have counteracted gravity induced flexion). However, due
to the insertion on the tibia, also the knee would be flexed
by the action of the BF. During hip extension during a
stride or a jump, it, therefore, seems reasonable to assume
that knee extensors are co-activated to prevent knee flexion.
In quantitative comparison, the two analysed specimens
of Cavia had an overall larger number of fascicles that
made up the BF than the analysed specimens of Chinchilla
(Table 2). However, despite considerable overlap in our
limited data set, fascicles of this muscle were much shorter
in Cavia than in Chinchilla in absolute (Table2) and rela-
tive terms (Fig. 2a). In both specimens of Cavia, the BF
had the largest number of fascicles of the three muscles
analysed. The specimens of Cavia had similar ACSA to
that of the analysed specimens of Chinchilla, but a much
smaller overall volume of the BF (Table 2). Pennation
angles were consistently larger relative to the x, y plane
than to the x, z plane in both species (Fig.2b, c). Absolute
values for PCSA and hence force-generating capacity were
similar in Cavia (CP1: PCSA = 2.34 cm², F = 52.65 N;
CP2: PCSA = 1.64 cm², F = 36.9 N) and Chinchilla (CC1:
PCSA = 1.97 cm²; F = 44.33 N; CC2: PCSA = 1.58 cm²,
F = 35.55N).
M. vastus lateralis (VL) In Cavia and Chinchilla the
VL is the largest of the distinct bellies of the quadriceps
femoris complex. In both species, it originated from the
Trochanter major and the Trochanter tertius and had a com-
mon insertion with the remaining bellies of the quadriceps
femoris via the patellar ligament. The ligament proximal to
the knee cap is very short in both species. Spanning solely
the knee joint, the muscle’s action would have extended the
knee (or would have counteracted gravity induced flexion)
during locomotor activity. Qualitatively, the muscle mass
of the VL appeared to be more concentrated proximally
and tapered considerably in Chinchilla in comparison with
Cavia which had a more evenly distributed mass along the
muscle belly.
The quantitative comparison of the VL revealed that
consistently across all analysed specimens this muscle was
made up of less fascicles and had a smaller ACSA and vol-
ume than both the BF and the TF (Table2). After correc-
tion for body mass and despite the considerable overlap, it
appears that the VL had relatively longer fascicles in the
two analysed specimens of Chinchilla than in the analysed
specimens of Cavia (Fig. 3a). In both species, pennation
angles relative to the x, y plane were considerably larger
than relative to the x, z plane (Fig. 3b, c). Approximated
PCSA and force-generating capacity of the VL were again
similar in Cavia (CP1: PCSA = 1.21cm², F = 27.23N; CP2:
PCSA = 0.69 cm², F = 15.53 N) than in Chinchilla (CC1:
PCSA = 1.34 cm², F = 30.15 N; CC2: PCSA = 0.96 cm²,
F = 21.6N).
M. triceps surae (TS) The TS consists of three parts: the
superficial medial and lateral gastrocnemius and the deep
Fig. 1 Topography of the three hind limb extensor muscles analysed
in this study. Left: Cavia porcellus, right: Chinchilla chinchilla. a, b
Photos of superficial muscles of the hind limb from lateral view to
show the M. biceps femoris (BF). c, d Photos of deeper muscles of
the hind limb to show the M. vastus lateralis (VL) and the M. triceps
surae (TS) after removal of the superficial BF. e, f: Renderings of the
digitised fascicles of the BF (red), VL (green), and fleshy (blue) and
tendinous (grey) fascicles of the TS for Cavia (E) and Chinchilla (F),
respectively
◂

Zoomorphology
1 3
soleus. The medial and lateral gastrocnemius originated
on the medial and lateral epicondyles of the femur, respec-
tively, as well as on the proximal head of the fibula. The
soleus also originated on the head of the fibula, but addi-
tionally had a large origin on the caudal aspect of the tibia
in both species. In Cavia and Chinchilla, the three parts of
the TS formed a single muscle belly and inserted via the
common Achilles tendon on the calcaneus. The Achil-
les tendon was considerably longer in Chinchilla than in
Cavia (Fig.1). Action of the TS would have plantar flexed
the ankle joint (or would have counteracted gravity induced
dorsiflexion) during locomotor bouts in both species. Fas-
cicles that originated on the distal femur, however, would
have also flexed the knee joint (albeit with a very unfavour-
able moment arm). Thus, co-activation of knee extensors
can be assumed during ankle extension. Individual fascicles
of the three parts of the TS could not be assigned to either
part during the quantitative analysis as all bellies appear to
fuse in deeper parts of the muscle.
Comparison of the quantitative architectural parameters
of the TS between both species demonstrates that this mus-
cle was made up of a similar number of fascicles. However,
in both analysed specimens of Chinchilla the TS consist-
ently comprised the most fascicles, because in this species,
the BF was made up of relatively fewer fascicles. In com-
parison with the analysed specimens of Cavia, both speci-
mens of Chinchilla also had longer fascicle lengths after
correction for body mass (Fig. 4a). In contrast, the ana-
lysed specimens of Cavia had consistently larger pennation
angles relative to both planes analysed here. Similar to the
VL, the TS in both species had larger pennation angles rel-
ative to the x, y plane than to the x, z plane (Fig.4b, c). In
both specimens of Chinchilla, the TS had the largest PCSA
and force-generating capacity of all three muscles analysed
in this study (CC1: PCSA = 2.33cm², F = 52.423 N; CC2:
PCSA = 1.72 cm², F = 38.7 N). In Cavia, consistently the
BF had a larger PCSA than the TS (CP1: PCSA = 1.81cm²,
F = 40.73N; CP2: PCSA = 1.15cm², F = 25.88N).
Discussion
In the current study, selected hind limb extensor muscles
of two closely related species of approximately the same
body size (Cavia and Chinchilla) were analysed to sub-
stantiate the notion that muscle architecture reflects adap-
tation in accordance with differing functional specialisa-
tions of the musculoskeletal system of mammals. Cavia
was also used in a hallmark study by Powell etal. (1984)
that experimentally verified the predictability of skeletal
muscle tension from muscle architectural properties. In
contrast to this previous study that used superficial meas-
urements to determine pennation angle and measured fibre
lengths in a limited sample of fibres, in the present study,
Table 2 Quantitative
architectural characteristics of
the BF, VL, and TS in Cavia
and Chinchilla
Data reported as mean (±standard deviation)
*Sum of the digitised fascicles from the left and right hindlimb of one specimen
**Arithmetic mean of the left and right hindlimb of one specimen
Fascicle length ACSA** Volume** Pennation angle
mm N* cm² mm³ ° to x, y plane ° to x, z plane
Cavia
CP1
Biceps femoris 7.81 (2.93) 1754 2.51 1939.29 24.47 (14.53) 17.53 (10.73)
Vastus lateralis 6.25 (2.35) 1058 1.30 811.63 28.59 (1.23) 14.06 (8.68)
Triceps surae 6.77 (2.64) 1157 1.91 919.68 22.09 (12.03) 14.24 (9.01)
CP2
Biceps femoris 10.45 (4.37) 1607 1.76 1751.91 26.08 (16.28) 16.25 (11.34)
Vastus lateralis 9.41 (3.46) 576 0.73 681.58 27.56 (13.07) 12.55 (7.62)
Triceps surae 8.90 (2.96) 829 1.20 757.87 20.89 (11.57) 10.74 (7.44)
Chinchilla
CC1
Biceps femoris 16.88 (7.11) 1255 2.11 4066.43 25.62 (14.37) 16.26 (12.95)
Vastus lateralis 11.70 (4.40) 1230 1.46 1773.15 33.81 (12.50) 11.76 (7.09)
Triceps surae 11.10 (3.91) 1596 2.41 1871.30 20.46 (13.57) 9.50 (6.83)
CC2
Biceps femoris 22.12 (10.21) 858 1.65 3648.54 22.66 (13.00) 16.22 (10.80)
Vastus lateralis 13.59 (4.81) 790 1.03 1408.92 30.61 (12.43) 10.43 (7.07)
Triceps surae 13.62 (4.70) 1053 1.77 1673.52 18.33 (12.61) 8.30 (6.18)

Zoomorphology
1 3
slightly larger pennation angles and considerably shorter
mean fascicle lengths were documented. Note that in the
current study fascicles, i.e,. bundles of approx. 5–50 fibres
(Lieber and Fridén 2001), instead of individual fibres were
digitised. Given that individual fibres often not even extend
the entire length of a fascicle (Loeb et al. 1987; Ounjian
et al. 1991), the current study probably will even overes-
timate the derived PCSAs. However, fascicle lengths and
pennation angles of the rabbit’s gastrocnemius and soleus
muscles measured by Siebert etal. (2015) using a similar
method as was used here found comparable values to those
that have been found here. Moreover, the method to deter-
mine architectural parameters used here was recently com-
pared to an automated method that uses diffusible iodine
contrast enhanced computed tomography (diceCT) and a
pattern recognition computer algorithm that detects fasci-
cles in computer tomography image stacks (Kupczik etal.
2015). Both methods identified similar fascicle lengths in
dog masseter muscles.
Even though the limited data set (only four BF, VL,
and TS per species) precludes data analysis using statisti-
cal methods, qualitative observations and quantitative data
appear to reflect locomotor specialisation of the specialised
jumper (Chinchilla) versus the striding quadruped (Cavia).
With one exception (see in the following), the expected dif-
ferences in the architectural properties were found. To aid
functional interpretation of these differences a performance
space plot was constructed (Fig. 5). For the performance
space plot, we pooled data of the BF, VL, and TS from the
two specimens belonging to the same species, respectively.
This can aid to detect trends, but nevertheless, and given
the considerable amount of variability within species in
our limited data set, more data are clearly necessary. Auto-
mated approaches using image data potentially allows for
the analysis of much larger data sets in future studies. Data
from just two specimens per species used in the current
study suggest that hind limb extensor muscles in Chinchilla
were relatively more powerful (plotted further towards the
upper right corner of the graph in Fig. 5) than in Cavia.
Moreover, functional interpretation of the data set assem-
bled here needs to be careful, because not all the extensors
of the studied joints have been analysed. Nevertheless, a
comparison of the architectural parameters of the homolo-
gous muscles in closely related species may yield tentative
insight into functional adaptation to differing functional
demands.
CC2CC1CP2CP1
BF fascicle length/body mas
s0.33
10
8
6
4
2
0
CC2CC1CP2CP1
BF pennation angle (°) to
x, y plane
30
25
20
15
10
CC2CC1CP2CP1
BF pennation angle (°) to
x, z plane
30
25
20
15
10
aivaCaivaCaivaC allihcnihCallihcnihCallihcnihC
CBA
Fig. 2 M. biceps femoris (BF) fascicle lengths distribution (a) and
pennation angles of fascicles relative to the muscle’s line of action in
the x, y (b) and x, z (c) planes. In a, all data have been corrected for
body mass, 50% of all values are within the boxes of the box-and-
whisker plots, the horizontal bar within the boxes represents the
median, each whisker represents 25% of the values, extreme values
are denoted as circles (at least 1.5 times the length of the box outside
of the box) and asterisks (at least 3 times the length of the box out-
side of the box). Error bars diagrams in b and c depict the mean and
95% confidence intervals. For better comparison b and c are shown in
the same scale as b and c in Figs.3 and 4, respectively

Zoomorphology
1 3
CC2CC1CP2CP1
10
8
6
4
2
0
VL pennation angle (°) to
x, z plane
CC2CC1CP2CP1
30
25
20
15
10
VL pennation angle (°) to
x, y plane
VL fascicle length/body mas
s0.66
Cavia Cavia CaviaChinchilla Chinchilla Chinchilla
ABC
CC2CC1CP2CP1
30
25
20
15
10
Fig. 3 Distribution of M. vastus lateralis (VL) fascicle lengths (cor-
rected for body mass) (a) and pennation angles of fascicles relative to
the muscle’s line of action in the x, y (b) and x, z (c) planes. Scale of
a as in Fig.2a to allow direct comparison. Box-and-whisker plots and
error bars defined as in Fig.2
TS pennation angle (°) to
x, y plane
TS fascicle length/body mass
0.66
Cavia Cavia CaviaChinchilla Chinchill
aC
hinchilla
ABC
TS pennation angle (°) to
x, z plane
CC2CC1CP2CP1
30
25
20
15
10
CC2CC1CP2CP1
10
8
6
4
2
0
CC2CC1CP2CP1
30
25
20
15
10
Fig. 4 Distribution of M. triceps surae (TS) fascicle lengths (cor-
rected for body mass) (a) and pennation angles of fascicles relative to
the muscle’s line of action in the x, y (b) and x, z (c) planes. Scale of
a as in Fig.2a to allow direct comparison. Box-and-whisker plots and
error bars defined as in Fig.2

Zoomorphology
1 3
Taken together, hind limb extensors of chinchillas
tended to have both, a relatively larger capacity for length
change and a greater capacity for force generation (but
again consider the limited sample). Similarly, Demes etal.
(1998) and Huq et al. (2015) found architectural proper-
ties that favour enhanced excursion and power in the hind
limb extensor muscles and epaxial muscles, respectively, of
specialised jumping primates when compared to primates
that were not specialised jumpers. As in these primates, the
caviomorph rodents analysed here differed in that the jump-
ing specialist also had relatively more voluminous muscles.
Given this limited sample, specialised jumpers, therefore,
appear to invest relatively more metabolic energy to build
up and maintain relatively larger muscles that are involved
in their specialised mode of locomotion.
The BF is the only exception to this pattern (and to
our expectation), because we found a greater relative
capacity for force generation in Cavia than in Chinchilla.
In Cavia, the BF was consistently the largest and most
forceful muscle. In rabbits (Oryctolagus cuniculus), the
BF was found to have by far the largest cross-sectional
area (and hence capacity for force generation) of all hind
limb muscles (Lieber and Blevins 1989), which, surpris-
ingly, was not the case in Chinchilla. In Chinchilla, the
TS was consistently larger than the BF (only four total
muscles analysed). Our finding of a greater capacity for
force production of the BF in the striding species might
be related to a proximo-distal gradient in joint control
during striding locomotion: movements of proximal
joints contribute the most to forward propulsion, whereas
distal limb joints mostly act to negotiate irregularities of
the ground (Kuznetsov 1985; Fischer etal. 2002; Daley
et al. 2007). Therefore, forceful hip extension might
be relatively more important for a striding quadruped
than for a specialised jumper which might explain the
presence of a more forceful hip extensor in Cavia than
in Chinchilla. It has also been argued that a need for a
short swing phase pendulum in species that use strid-
ing locomotion constrains mass distribution along the
leg and leads to a proximal concentration of mass (e.g.,
Demes et al. 1998; Kilbourne et al. 2016). However,
while appearing to have slightly lesser force-generating
capacity (in the pooled data of Fig. 5, but note that no
clear trend can be determined from our limited data set),
the BF of Chinchilla had by far the greatest capacity for
length change of all analysed muscles due to its consist-
ently long fascicles (across all specimens of our data set,
Chinchilla had absolutely and relatively longer BF fasci-
cles in the mean). This can be argued to reflect the need
for extensive excursions of the hip when jumping out of a
crouched posture (Legreneur etal. 2010).
During symmetrical gaits of small to medium sized
mammals the knee has been demonstrated to not undergo
large excursions (e.g., Fischer etal. 2002; Rocha-Barbosa
et al. 2005; Fischer and Blickhan 2006). During a stride,
activity of knee extensors like the VL, therefore, can be
expected to mainly prevent the knee from gravity induced
flexion, i.e., they fulfil mostly an anti-gravity function.
Additionally, during striding locomotion, knee extensors
are required to counteract knee flexion induced by activ-
ity of bi-articular muscles like the BF and gastrocnemius
(i.e., the part of the TS that originates from the femoral
condyles). This is, because the activity of the latter muscles
would not only extend the hip and ankle, respectively, but
also flex the knee. This somewhat limited functional role
of knee extensors during symmetrical gaits in addition to
a constrained proximo-distal mass distribution (cf. Demes
etal. 1998) might explain the relatively small capacity to
perform work of the VL in Cavia. However, in jumping
mammals powerful extension of the more distal hind limb
joints (knee and ankle) appears to be of paramount impor-
tance (e.g., Demes etal. 1996; Aerts 1998, Essner Jr. 2002;
Legreneur et al. 2010). In agreement with this, relatively
more powerful knee extensors, as found here in Chinchilla,
have also been found in other specialised jumpers in a com-
parison of a sample of quadruped primates and vertical
clinging and leaping primates (Demes etal. 1998). While
vertical clinging and leaping involves different kinematics
due to the vertical launching support, powerful knee exten-
sion nevertheless seems to be important for both jumping
behaviours (horizontal and vertical launches).
0
0.01
0.02
0.03
0.04
0.05
00.5 11.5 22.5 3
fascicle length/body mass
0.33
ssamydob/ASCP 0.66
VL
VL
BF
BF TS
TS
‘lufrewop‘‘lufecrof‘
‘length change‘‘general‘
Fig. 5 Performance space plot of analysed hind limb extensor mus-
cles (BF biceps femoris, VL vastus lateralis, TS fleshy fascicles of the
triceps surae) in Cavia (squares) and Chinchilla (circles) of normal-
ised fascicle length versus normalised PCSA (pooled data for each
species). This graph can be used to compare the relative forces and
excursions of the analysed muscles (cf. Lieber and Fridén 2001).
‘Forceful’ muscles have relatively large PCSA, whereas muscles spe-
cialised for ‘length change’ have relatively long fascicles. ‘Powerful’
muscles are characterised by both, relatively large PCSA and rela-
tively long fascicles. However, these are metabolically more expen-
sive. To account for the slight differences in body mass, all data have
been normalised according to geometric scaling (cf. Allen etal. 2010;
Dick and Clemente 2016)

Zoomorphology
1 3
Finally, the ankle extensor (TS) again differed in accord-
ance with the expectations between the analysed specimens
of the studied quadruped (Cavia) and the jumper (Chin-
chilla) in this study. Not only did this muscle reflect a spe-
cialisation towards more powerful extension of the ankle
in Chinchilla, but also this muscle was the most forceful
muscle of our sample relative to body mass (in contrast to
rabbits, cf. Lieber and Blevins 1989). In addition, the ten-
dinous part was by far longer in Chinchilla than in Cavia.
Complex interactions between the muscle and its tendon
may change from low power activities like slow hopping
to high power activities like rapid accelerations (Roberts
2002). Especially, during high power activities, work from
muscles can be transferred to the tendon which is initially
stretched and then redistributes muscle power over the
duration of the activity, e.g., the launch. Tendon stretch and
recoil thus extends the functional range of muscles (Rob-
erts 2002). Apparently, the jumping species in our study
has increased capability to store and release elastic strain
energy (Alexander and Bennet-Clark 1977; Roberts 2002).
In conclusion, the qualitative and quantitative compari-
son of architectural characteristics of selected hind limb
extensor muscles in two specimens of Cavia and Chin-
chilla, respectively, highlights adaptations of muscle archi-
tecture in the specialised jumping species. As expected,
the hind limb extensors of Chinchilla were relatively more
powerful with greater capacity for length change and force
generation (Fig. 5), even though we also found a consid-
erable variability between the specimens of our study that
accounted for the heterogeneity of architectural parameters
within the muscles (Table 2). Alternatively, it is possible
that this variability arises from the fact that all data of the
current study originated from captive animals from breed-
ers and a limited sample size. Nevertheless, these results
tentatively add to the overall notion that muscle architec-
ture is reflecting differing functional demands of muscles
within the same organism (e.g., Lieber and Blevin 1989;
Payne et al. 2005; Stark et al. 2013; Moore et al. 2013;
Rupert etal. 2015; Nyakatura and Stark 2015), is adjusted
to changing functional demands inflicted by increasing
body size during ontogeny (e.g., Allen etal. 2010), and also
is reflecting differing functional demands of closely related
species with different locomotor behaviours (cf. Payne
etal. 2006; Allen etal. 2014; Huq etal. 2015) or body size
(Dick and Clemente 2016). Muscle architecture, therefore,
further proves to present an informative link between form
(e.g., bone morphology with muscle attachment sites and
lever arms) and function (e.g., invivo motion analyses and
behavioural observations) in the context of biological adap-
tation to differing functional demands.
Acknowledgements The authors thank Heiko Stark for his help
during various stages of this study. This study received funding from
the Deutsche Forschungsgemeinschaft (DFG), Grant No. EXC 1027
“Bild Wissen Gestaltung: ein interdisziplinäres Labor”.
Compliance with ethical standards
All applicable international, national, and/or institutional guidelines
for the care and use of animals were followed. This article does not
contain any studies with human participants performed by any of the
authors.
Conflict of interest The authors declare that they have no conflict
of interest.
References
Aerts P (1998) Vertical jumping in Galago senegalensis: the quest for
an obligate mechanical power amplifier. Phil Trans Roy Soc B
Biol Sci 353(1375):1607–1620
Alexander RM, Bennet-Clark HC (1977) Storage of elastic strain
energy in muscle and other tissues. Nature 265(5590):114–117
Allen V, Elsey RM, Jones N, Wright J, Hutchinson JR (2010) Func-
tional specialization and ontogenetic scaling of limb anatomy in
Alligator mississippiensis. J Anat 216(4):423–445
Allen V, Molnar J, Parker W, Pollard A, Nolan G, Hutchinson JR
(2014) Comparative architectural properties of limb muscles
in Crocodylidae and Alligatoridae and their relevance to diver-
gent use of asymmetrical gaits in extant Crocodylia. J Anat
225(6):569–582
Biewener AA (2003) Animal locomotion. Oxford University Press,
New York
Close RI (1972) The relations between sarcomere length and charac-
teristics of isometric twitch contractions of frog sartorius mus-
cle. J Physiol 220(3):745
Daley MA, Felix G, Biewener AA (2007) Running stability is
enhanced by a proximo-distal gradient in joint neuromechanical
control. J Exp Biol 210(3):383–394
Demes B, Jungers WL, Fleagle JG, Wunderlich RE, Richmond BG,
Lemelin P (1996) Body size and leaping kinematics in Malagasy
vertical clingers and leapers. J Hum Evol 31(4):367–388
Demes B, Fleagle JG, Lemelin P (1998) Myological correlates of pro-
simian leaping. J Hum Evol 34(4):385–399
Demes B, Franz TM, Carlson KJ (2005) External forces on the limbs
of jumping lemurs at takeoff and landing. Am J Phys Anthropol
128(2):348–358
Dick TJ, Clemente CJ (2016) How to build your dragon: scal-
ing of muscle architecture from the world’s smallest to the
world’s largest monitor lizard. Front Zool 13(1):1. doi:10.1186/
s12983-01-0141-5
Dumas GA, Poulin MJ, Roy B, Gagnon M, Jovanovic M (1988) A
three-dimensional digitization method to measure trunk muscle
lines of action. Spine 13:532–541
Elissamburu A, Vizcaíno SF (2004) Limb proportions and adapta-
tions in caviomorph rodents (Rodentia: Caviomorpha). J Zool
262(02):145–159
Epstein M, Herzog W (1998) Theoretical models of skeletal muscle:
biological and mathematical considerations. Wiley, Chichester
Essner RL (2002) Three-dimensional launch kinematics in leaping,
parachuting and gliding squirrels. J Exp Biol 205(16):2469–2477
Fischer MS, Blickhan R (2006) The tri-segmented limbs of therian
mammals: kinematics, dynamics, and self-stabilization—a
review. J Exp Zool A Comp Exp Zool 305(11):935–952
Fischer MS, Schilling N, Schmidt M, Haarhaus D, Witte H (2002)
Basic limb kinematics of small therian mammals. J Exp Biol
205(9):1315–1338

Zoomorphology
1 3
Fukunaga T, Ichinose Y, Ito M, Kawakami Y, Fukashiro S (1997)
Determination of fascicle length and pennation in a contracting
human muscle invivo. J Appl Physiol 82(1):354–358
Gans C, Bock WJ (1964) The functional significance of muscle
architecture—a theoretical analysis. Ergebnisse der Anatomie
Entwicklungsgeschichte 38:115–142
Huq E, Wall CE, Taylor AB (2015) Epaxial muscle fiber architecture
favors enhanced excursion and power in the leaper Galago sen-
egalensis. J Anat 227(4):524–540
Kilbourne BM, Andrada E, Fischer MS, Nyakatura JA (2016) Mor-
phology and motion: hindlimb proportions and swing phase
kinematics in terrestrially locomoting charadriiform birds. J Exp
Biol 219(9):1405–1416
Kim SY, Boynton EL, Ravichandiran K, Fung LY, Bleakney R, Agur
AM (2007) Three-dimensional study of the musculotendinous
architecture of supraspinatus and its functional correlations. Clin
Anat 20(6):648–655
Kupczik K, Stark H, Mundry R, Neininger FT, Heidlauf T, Röhrle
O (2015) Reconstruction of muscle fascicle architecture from
iodine-enhanced microCT images: a combined texture mapping
and streamline approach. J Theor Biol 382:34–43
Kuznetsov AN (1985) Comparative functional analysis of the fore-
and hind limbs in mammals. Zool J Moscow 64:1862–1867
Legreneur P, Thévenet FR, Libourel PA, Monteil KM, Montuelle S,
Pouydebat E, Bels V (2010) Hindlimb interarticular coordina-
tions in Microcebus murinus in maximal leaping. J Exp Biol
213(8):1320–1327
Lieber RL, Blevins FT (1989) Skeletal muscle architecture of the rab-
bit hindlimb: functional implications of muscle design. J Morph
199(1):93–101
Lieber RL, Fridén J (2001) Clinical significance of skeletal muscle
architecture. Clin Orthop Rel Res 383:140–151
Loeb GE, Pratt CA, Chanaud CM, Richmond FJR (1987) Distribution
and innervation of short, interdigitated muscle fibers in parallel-
fibered muscles of the cat hindlimb. J Morph 191(1):1–15
Moore AL, Budny JE, Russell AP, Butcher MT (2013) Architectural
specialization of the intrinsic thoracic limb musculature of the
American badger (Taxidea taxus). J Morph 274(1):35–48
Nowak RM (1999) Walker’s mammals of the world, 6thedn. Johns
Hopkins University Press, Baltimore
Nyakatura JA, Stark H (2015) Aberrant back muscle function corre-
lates with intramuscular architecture of dorsovertebral muscles
in two-toed sloths. Mamm Biol 80(2):114–121
Ounjian M, Roy RR, Eldred E, Garfinkel A, Payne JR, Armstrong A,
Edgerton VR (1991) Physiological and developmental implica-
tions of motor unit anatomy. J Neurobiol 22(5):547–559
Payne RC, Hutchinson JR, Robilliard JJ, Smith NC, Wilson AM
(2005) Functional specialisation of pelvic limb anatomy in
horses (Equus caballus). J Anat 206(6):557–574
Payne RC, Crompton RH, Isler K, Savage R, Vereecke EE, Günther
MM, D’Août K (2006) Morphological analysis of the hindlimb in
apes and humans. I. Muscle architecture. J Anat 208(6):709–724
Poelstra KA, Eijkelkamp MF, Veldhuizen AG (2000) The geometry
of the human paraspinal muscles with the aid of three-dimen-
sional computed tomography scans and 3-Space Isotrak. Spine
25(17):2176–2179
Powell PL, Roy RR, Kanim PAULA, Bello MA, Edgerton VR (1984)
Predictability of skeletal muscle tension from architectural
determinations in guinea pig hindlimbs. J Appl Physiol
57(6):1715–1721
Roach N, Kennerly R (2016) Chinchilla chinchilla. The IUCN Red
List of Threatened Species (e.T4651A22191157)
Roberts TJ (2002) The integrated function of muscles and tendons
during locomotion. Comp Biochem Physiol A Mol Integr Phys-
iol 133(4):1087–1099
Rocha-Barbosa O, De Castro Loguercio MF, Renous S, Gasc JP
(2005) Limb joints kinematics and their relation to increasing
speed in the guinea pig Cavia porcellus (Mammalia: Rodentia). J
Zool 266(3):293–306
Rosatelli AL, Ravichandiran K, Agur AM (2008) Three-dimensional
study of the musculotendinous architecture of lumbar multifidus
and its functional implications. Clin Anat 21(6):539–546
Rupert JE, Rose JA, Organ JM, Butcher MT (2015) Forelimb muscle
architecture and myosin isoform composition in the groundhog
(Marmota monax). J Exp Biol 218(2):194–205
Sacks RD, Roy RR (1982) Architecture of the hind limb muscles of
cats: functional significance. J Morph 173(2):185–195
Scholle HC, Schumann NP, Biedermann F, Stegeman DF, Grassme
R, Roeleveld K, Schilling N, Fischer MS (2001) Spatiotemporal
surface EMG characteristics from rat triceps brachii muscle dur-
ing treadmill locomotion indicate selective recruitment of func-
tionally distinct muscle regions. Exp Brain Res 138(1):26–36
Siebert T, Leichsenring K, Rode C, Wick C, Stutzig N, Schubert
H, Böl M (2015). Three-dimensional muscle architecture and
comprehensive dynamic properties of rabbit gastrocnemius,
plantaris and soleus: input for simulation studies. PLoS one
10(6):e0130985
Spector SA, Gardiner PF, Zernicke RF, Roy RR, Edgerton VR (1980)
Muscle architecture and force-velocity characteristics of cat
soleus and medial gastrocnemius: implications for motor control.
J Neurophysiol 44(5):951–960
Spotorno AE, Zuleta CA, Valladares JP, Deane AL, Jiménez JE
(2004) Chinchilla laniger. Mamm Spec 758:1–9
Spotorno AE, Manríquez G, Fernández LA, Marín JC, González F,
Wheeler J (2007) Domestication of guinea pigs from a southern
Peru-northern Chile wild species and their middle pre-Colum-
bian mummies. In: Kelt DA, Lessa EP, Salazar-Bravo J, Patton
JL (eds) The Quintessential Naturalist: Honoring the Life and
Legacy of Oliver P. Pearson. University of California Publica-
tions in Zoology, pp367–388
Stark H, Fröber R, Schilling N (2013) Intramuscular architec-
ture of the autochthonous back muscles in humans. J Anat
222(2):214–222
Wells JB (1965) Comparison of mechanical properties between slow
and fast mammalian muscles. J Physiol 178(2):252
Wickiewicz TL, Roy RR, Powell PL, Edgerton VR (1983) Mus-
cle architecture of the human lower limb. Clin Orthop Rel Res
179:275–283
Wickiewicz TL, Roy RR, Powell PL, Perrine JJ, Edgerton VR (1984)
Muscle architecture and force-velocity relationships in humans. J
Appl Physiol 57(2):435–443
Zajac FE (1989) Muscle and tendon Properties models scaling and
application to biomechanics and motor. Crit Rev Biomed Eng
17(4):359–411