Comparative Analysis of Paw Pad Structure in the Clouded Leopard (Neofelis nebulosa) and Domestic Cat (Felis catus)

Authors


Abstract

The Clouded leopard (Neofelis nebulosa) is a medium-sized highly arboreal cat. This study compares the structure of the digital, metacarpal and metatarsal pads of the manus and pes in N. nebulosa to that of the domestic cat (Felis catus). Covered by a stratified squamous cornified epithelium, the pads have a supple deposit of subepidermal fat that is partitioned by collagen fibers and extensively anchored to the muscle tendon sheaths. In both animals, a pes metatarsal pad suspensory ligament originates from the Mm. flexores digitorum profundi tendon and forms 3–4 small branches that project through the dermal fat layer and attach to the pad epidermis. In the cat manus, four tendons of equal size extend from the M. flexor digitorum superficialis (FDS) to form the manica flexoria in digits 2–4 from which extends a metacarpal pad suspensory ligament (MPSL) on digits 2 and 5 that extends into the tela subcutanea and epidermis. On digits 3 and 4 MPSL extends directly from the FDS tendon itself. In contrast, manus FDS tendons 1 and 5 in N. nebulosa were thin and either project directly to the tela subcutanea (tendon 1) or connect with the manica flexoria forming a metacarpal pad suspensory ligament (tendon 5). Tendons 2–4 connect with the manica flexoria from which MPSL project into the tela subcutanea and epidermis. In both species, the suspensory ligaments may serve to contract the pad to conform to the under lying substrate, thus enhancing the animal's ability to grip branches while climbing. Anat Rec, 2009. © 2009 Wiley-Liss, Inc.

The Clouded leopard (Neofelis,nebulosa, N. nebulosa) is the smallest “big cat,” that has the body size and shape of a small cat but with pantherine cranial and dental structure. These cats have the longest canine tooth-to-body size ratio among extant felids (Nowell and Jackson, 1996; Sunquist and Sunquist, 2002), and canines with sharp-edged posterior keels, similar to extinct saber-tooth cats (Christiansen, 2006). Canine length affects other cranial features, including muzzle elongation, a lengthened diastema between the canine and reduced first premolar, and a greatly elongated skull (Guggisberg, 1975). These cats also have short legs, with even shorter forelegs, robust bodies and greatly enlarged forefeet that can assume a spoon shape for more effective gripping of supporting branches. N. nebulosa weigh from 16 to 23 kg and are about 60–106-cm long (head and body). The extremely long tail, used for balancing while climbing, adds 55–91 cm. to their total length (Wozencraft, 2005). N. nebulosa are solitary except for females with young (Yamada and Durant, 1989).

The common name, Clouded leopard, derives from coat markings; large, irregular cloud-like pale spots outlined in black or brown surrounded by a pale ring on pale yellow to light brown background (Gunderson, 1976; Nowell and Jackson, 1996; Sunquist and Sunquist, 2002). Recent coat color, pattern variation and genetic analyses divide N. nebulosa into two species (Grassman et al., 2005; Buckley-Beason et al., 2006; Kitchener et al., 2006). N. nebulosa occurs widely across mainland Asia (Loxton, 1973), while the new N. diardi is restricted to Indonesia (Kitchener et al., 2006). Both species occur in mangrove swamps in Borneo (Davis, 1962; Davies and Payne, 1982), grass and scrublands (Santiapillai and Ashby, 1988; Johns, 1989; Dinerstein and Mehta, 1989; Rabinowitz and Walker, 1991) subtropical forests, primary closed evergreen tropical forests, (Rabinowitz, 1988; Austin, 2002), tropical rainforests (Johns, 1989) and open, dry partially forested areas. N. nebulosa adapt to different habitats, and is not restricted by human perturbation, because they can occupy semidisturbed habitats (Rabinowitz et al., 1987) of up to 3,000 square meters (Jerdon, 1874; Biswas et al., 1985). Habitat destruction is a major contributor to their highly endangered status [(CITES I); Groombridge, 1994; Baillie and Groombridge, 1996; Nowell and Jackson, 1996] as is poaching for Asian medicines and their beautiful coats (Low, 1991).

N. nebulosa are rare, shy, mainly nocturnal, and arboreal, hunting and resting in trees (Grassman, 2001). They are generalist, opportunistic predators, taking birds, arboreal monkeys and large prey, including pigs, cattle, young buffalo, goats, deer, and even porcupines (Nowak, 1999). These cats differ from typical felids in being able to run down trees headfirst, and to move along horizontal supports with the body suspended beneath the branches (Layhausen, 1979; Choudhury, 1996; Conforti, 1996; Nowell and Jackson, 1996; Ghose, 2002; Grassman and Tewes, 2002). These unusual, highly acrobatic abilities have been also reported in the much smaller margay and marbled cats, but the larger N. nebulosa displays greater adaptations for extreme arboreal abilities (Leopold, 1959; Grzimek, 1975; Lekagul and McNeely, 1977; Nowell and Jackson, 1996). The margay (Leopardusweidii) is from Central and South America, and the marbled cat, (Pardofelismarmorata) shares the Asian range of N. nebulosa (Pocock, 1932; Sunquist and Sunquist, 2002). The margay and perhaps the marbled cat, have increased intratarsal flexibility to allow movements that are impossible in other felids (Grzimek, 1975), but there are no studies of more superficial structures that also correlate with tarsal movements (Layhausen, 1979; Choudhury, 1996; Ghose, 2002). Information on gross and histological anatomy and function of paw pads is sparse, and unsurprisingly, nothing has been published on these structures in N. nebulosa (Cutts and Krause, 1983; Alexander et al., 1986; Benz et al., 2005.).

MATERIALS AND METHODS

Two N. nebulosa were obtained through the Smithsonian Institution Osteopreparatory laboratory and are housed in the research collection of the biological sciences department of Northern Illinois University (NIU 2105-2106). Both animals died of natural causes. Following necropsy, the animals were transported to Northern Illinois University for further dissection and analysis of the recorded results. Preserved specimens of six, F. catus, were obtained from NASCO Biological Supply Company and were studied as a comparison to N. nebulosa. Before dissection, tissue sample cubes were removed for histological analysis from the metacarpal and metatarsal pads (∼10 mm2 for N. nebulosa and 5 mm2 for the domestic cat) as well as the digital pads (∼5 mm2 for N. nebulosa and 2–3 mm2 for F. catus). Each sample of tissue included epidermis, dermis and subdermis, ending at the deep fascia covering the muscle or tendons of the metacarpal or metatarsal pads. One sample block was taken from of the center and one from the medial and lateral side of the metacarpal or metatarsal pads of manus and pes for each N. nebulosa. A single sample was removed from the center of the fore and hind paw pads of two F. catus specimens. One sample was obtained from each of digits 2–4 of each manus and pes. Samples were placed in buffered formalin for 1 week then dehydrated and embedded in paraffin. The tissue blocks were sectioned yielding 10–15 serial sections (depending on the size) which were placed on a glass slide for staining using standard histological methods (Presnell and Schreibman, 1997). Three slides from each sample block were chosen randomly. One each was stained with either hemotoxylin and eosin (H&E, Presnell and Schreibman, 1997), a gold nerve stain (Garvey et al., 1987) or Verhoeff's elastin stain (Presnell and Schreibman, 1997). A representative section was chosen at random in each H&E slide. The stratium cornium layer and combined noncornified layers were then measured separately at 40× for the N. nebulosa or at 100× for F. catus using an ocular micrometer (calibrated at 40× and 100× power using the 1/10 mm etched squares in a hemocytometer). The relative adipose content in various layers of the sample, nerve fiber numbers and location, and elastic fiber density were noted for each slide.

Radiographic images of N. nebulosa and F. catus were used to compare the relative size of the manus and pes. Radiographs were obtained using a portable General Electric Mobile 100-15 X-ray machine and AXB-90 Alpha Images Medical high speed film. For each specimen, each paw was X-rayed separately in an anterior/posterior position. Following film development the radiographs were analyzed by measuring the distance between the lateral boarders of digits 2 and 5 with a line that passed between the metacarpal/metatarsal phalangeal joints of digits 3 and 4 (line 1, Fig. 1). A second measurement was made of the length of the third metacarpal bone (line 2, Fig. 1). These two measurements were then expressed as a ratio (width/length) and averaged for the manus and pes of each animal.

Figure 1.

Illustration of the method used to measure manus size ratio. The radiograph shown is that of a N. nebulosa manus. The length of a straight line drawn between the fifth and second proximal phalanx parallel to the third and fourth metacarpalphalangeal joint was determined (1). This value was divided by the length of the third metacarpal (2) to yield a width/length ratio.

RESULTS

Anatomy of Manus and Pes Metacarpal, Metatarsal and Digital Pads

F. catus

Manus.

Although the anatomy of F. catus has been more completely documented than that of many domestic species, many aspects remain to be investigated (Reighart and Jennings, 1925; Crouch, 1969; Walker, 1986; Rosenzweig, 1990). In particular, we were interested in studying the structural features of the metacarpal, metatarsal and digital pads in the manus and pes that may assist in locomotor activities. The most notable of our findings was the presence of a group of small ligamentous structures that are collectively designated in this article as either metacarpal (manus) or metatarsal (pes) pad suspensory ligaments. These structures project into the connective tissue of the metacarpal/metatarsal pad and may also be attached to the skin (Fig. 2). In addition, when referring to characteristics common to both the metacarpal and metatarsal pad the term “paw pad” will be used. Digitial pads refer to pads associated with the digits.

Figure 2.

This illustration is a composite diagram of gross and histologic structures observed in the manus of N. nebulosa. (1) Metacarpal suspensory ligaments project from the M. flexor digitorum superficialis tendons of digits 3, 4, and 5 into the metacarpal fat pad and appear to attach to regions adjacent to the epidermis. (2) Metacarpal pad suspensory ligaments extend from the manica flexoria of digit 2 into the metacarpal fat pad. (3) Metacarpal pad suspensory ligaments from the distal end of manica flexoria of digits 2–4 into the metacarpal pad connective tissue. Bundles of collagen fibers running parallel to the epidermis and adipose tissue compartments are present in the subcutaneous compartment.

In F. catus, M. flexor digitorum superficialis originates on the medial epicondyle of the humerus. This muscle along with M. interflexoria and M. flexor digitorum brevis join to form a single flattened tendon that passes deep to the flexor retinaculum where it divides into five smaller tendons that extend to each of the five digits (Fig. 3). Tendons to digits 2–5 each form the manica flexoria fibrous collars through which the Mm. flexor digitorum profundus tendons pass on the way to the distal phalanges (Waibl et al., 2005). Each of the M. flexor digitorum superficialis tendons to digits 2–5 pass deep to the transverse metacarpal ligament at which point tendons to digits 3 and 4 each give off a single metacarpal pad suspensory ligament that passes into the connective tissue of the metacarpal pad (Figs. 3, 4). These two branches attach to the skin of the metacarpal pad. Extending from the palmar surface of the manica flexoria on digits 2 and 5 are short metacarpal pad suspensory ligaments that project into the tela subcutanea of the metacarpal pad (not shown). In addition numerous small fibers were seen running in the metcarpal pad region from the deep fascia toward the epidermis (not shown). No metacarpal pad suspensory ligaments were seen associated with the flexor tendons to the pollex.

Figure 3.

The left manus of F. catus with a majority of the metacarpal pad removed shows four tendons from M. flexor digitiorum superficialis to digits 2–5 that run deep to the metacarpal fat pad and pass beneath the transverse metacarpal ligament. Metacarpal pad suspensory ligaments to digits 3 and 4 can be seen projecting into the distal region of the metacarpal fat pad. M. interflexoria tendon to digit 2 can be seen deep to the M. flexor digitorum superficialis tendon. Number 1 marks the first digit.

Figure 4.

The left manus of F. catus viewed from the opposite side (rotated 180°) when compared with Fig. 1 (see pollex marked 1). The transverse metacarpal ligament has been removed, showing the tendons of M. flexor digitiorum superficialis to digits 2–5, M. flexor digitorum brevis tendons to digits 4 and 5 and the metacarpal pad suspensory ligaments to digits 3 and 4 (elevated by the forceps). Number 1 marks the first digit.

Pes.

The metatarsal pad suspensory ligament (also called common plantar ligament but will be designated as the metatarsal pad suspensory ligament in this article) has been previously described for F. catus (Crouch, 1969; Walker, 1986; Rosenzweig, 1990). Mm. flexores digitorum profundi originate in the leg as three muscles (M. flexor digitorum lateralis, M. flexor digitorum medialis and M. tibialis caudalis). M. Flexor digitorum lateralis and M. flexor digitorum medialis unite to form a single tendon that splits into four tendons in the metatarsal region each of which inserts onto a distal phalanx. M. interflexoria is attached to the plantar side of the Mm. flexores digitorum profundi tendon expansion in the metatarsal region of the pes. M. interflexoria forms three tendons that join those of M. flexor digitorum superficialis going to digits 3–5. A metatarsal pad suspensory ligament projects from the center of the tendon expansion of Mm. flexores digitorum profundi, piercing M. interflexoria, M. flexor digitorum superficialis, and M. flexor digitorum brevis (Fig. 5). The metatarsal pad suspensory ligament then splits into three branches that spread in a fan-shaped pattern through the connective tissue of the metatarsal pad. Close inspection with a dissecting scope (30×) showed that the final attachment site for each branch was to the skin of the metatarsal pad.

Figure 5.

A: The right pes of F. catus showing the metatarsal pad suspensory ligament projecting through M. flexor digitorum brevis into the metatarsal pad connective tissue. B: The metatarsal pad, M. flexor digitorum brevis and M. interflexoria and have been removed showing the metatarsal pad suspensory ligament attached to the Mm. flexores digitorum profundi tendon.

N. nebulosa

Manus.

In the N. nebulosa, the M. flexor digitorum superficialis tendon projects into the metacarpal region of the manus as a thick rounded tendon that forms five branches, one to each digit (Figs. 6, 7). Tendons to digits 1 and 5 are thin and either extend directly from the central branching point of M. flexor digitorum superficialis to the connective tissue of the metacarpal pad (digit 1) without contacting the manica flexoria or have a small thin connection to the manica flexoria in addition to a branch that extends into the connective tissue of the metacarpal pad (digit 5, not clearly visible, Figs. 6B, 7A). Digit 2 has a small short metacarpal pad suspensory ligament projecting from the manica flexoria into the tela subcutanea of the metacarpal pad (not shown). On digits 3 and 4, a small single metacarpal pad suspensory ligament projects from each tendon at a point ∼1 cm. from the branch point where M. flexor digitorum superficialis tendon splits into 5 tendons (Fig. 7A,B). Examination of the metacarpal pad suspensory ligament to digits 3, 4, and 5 using a dissecting microscope showed that each branch fanned out at the distal end and attached to the surrounding connective tissue as well as the dermis immediately adjacent to the epidermis of the pad (not shown). Further examination of the manica flexoria distal to the attachment of the M. flexor digitorum superficialis on digit 2–4 showed short metacarpal pad suspensory ligaments extending from the manica flexoria into the surrounding tela subcutanea of the metacarpal pad (illustrated in Figs. 2, 7). As with F. catus, the metacarpal pad region of the manus in the N. nebulosa had numerous small fibers running from the deep fascia toward the epidermis.

Figure 6.

A: The left manus of N. nebulosa showing M. flexor digitorum superficialis tendon as it passes from beneath M. flexor digitorum brevis and then branches into five tendons. B: The M. flexor digitorum superficialis tendon has been separated from the underlying tissue and M. flexor digitorum brevis. The digit 1 in (A) and (B) is marked with the number 1.

Figure 7.

A: The left manus of N. nebulosa showing the M. flexor digitorum superficialis tendon. Most of the metacarpal pad has been removed to reveal the tendons. Metacarpal pad suspensory ligaments can be seen projecting from M. flexor digitorum superficialis tendons 2, 3, and 4. Digit 1 is marked with the number 1. B: A closer view of flexor tendons (M. flexor digitorum superficialis, M. interflexoria and M. flexor digitorum profunda) to digit 4 and its attendant metacarpal pad suspensory ligament. M. flexor digitorum superficialis ligament to digit 5 can also be seen marked with an asterisk (*) at point where it splits forming two branches; one to the manica flexoria and the other to the metacarpal pad connective tissue.

Pes.

The metatarsal pad suspensory ligament in the N. nebulosa projects from the expansion of the Mm. flexores digitorum profundi tendon. It pierces M. interflexoria then runs between tendons 3 and 4 of M. flexor digitorum superficialis and deep to M. flexor digitorum brevis on its way to the tela subcutanea of the metatarsal pad (Figs. 8, 9). This ligament forms three branches that pass through the connective tissue and attach to the skin of the medial, lateral and central regions of the metatarsal pad (Fig. 8B).

Figure 8.

A: The right pes of N. nebulosa is shown in posteroventral view. The skin has been removed revealing the M. flexor digitorum brevis. B: The distal portion of M. flexor digitorum brevis and the central tendons of M. flexor digitorum superficialis have been removed showing the metatarsal pad suspensory ligament projecting into the metatarsal pad. Four branches can be seen projecting from the metatarsal pad suspensory ligament that insert into the subcutaneous tissue of the metatarsal pad.

Figure 9.

A deep dissection of the right pes of N. nebulosa showing the Mm. flexores digitorum profundi tendon and the attachment site of the metatarsal pad suspensory ligament. The M. flexor digitorum brevis and the M. flexor digitorum superficialis tendon have been detached and folded toward the distal end of the pes. Mm. interflexorii has been removed to show the attachment site of the metatarsal pad suspensory ligament. M. quadratus plantae can also be seen attached to Mm. flexores digitorum profundi tendon.

In addition to the tendons described earlier, numerous small anchoring fibers were seen running through the adipose connective tissue of the metatarsal pad in both the manus and pes (not shown). These fibers appear to be attached to the tendon sheaths deep to the metatarsal pad. They pass through the connective tissue and attach to the skin of the metatarsal pad.

Histological Analysis

The central thickness of the metacarpal and metatarsal pads extending from the muscle tendons to the epidermal surface in F. catus averaged 4 mm (3.4–4.5 mm). Central digital pad thickness averaged 2 mm (1.25–2.5 mm).

In the N. nebulosa, the metacarpal and metatarsal pad thickness were each ∼10 mm (8–12 mm) while the digital pads were ∼4 mm (3.4–4.5 mm) at the thickest point.

Tissue samples stained with H&E from F. catus and N. nebulosa showed similar morphologic characteristics except for the epithelial thickness. Initial analysis of sample measurements taken from the medial, lateral, and central metacarpal and metatarsal pads in the N. nebulosa and F. catus were not significantly different from each for a given pad, so all three values were averaged to yield a single value for each paw (Table 1). For the N. nebulosa, the thickness of the stratum cornium and the noncornified epithelial layers in the metacarpal/metatarsal pads were significantly greater than the same layers in the digital pads (P ≤ 0.05). In F. catus, the thickness of the metacarpal/metatarsal pad stratum cornium was also greater than the stratum cornium in the digital pads but the noncornified layer was not different for both tissues. Tall dermal papillae projected from the dermis into the stratum basalis of the epidermis (Fig. 10). Deep to the epidermis were thick bundles of collagen fibers running parallel to the epidermis (Fig. 10). Individual collagen septae passed from the subcutaneous regions of the metacarpal pad toward the epidermis. The septal fibers appeared to interdigitate at perpendicular or oblique angles with the parallel subepidermal collagen fibers bundles. Adipose tissue was segregated by connective tissue into small individual compartments throughout the hypodermis. Nerves and blood vessels were present throughout the connective tissue in the hypodermis along with eccrine sweat glands, and sweat gland ducts that ran between the adipose tissue compartments.

Table 1. The thickness of the stratium cornium compared to non-cornified layers of the epidermis in the metatarsal, metacarpal and digital pads of the N. nebulosa and domestic cat
  1. Values are expressed as mean μm ± SEM. For both N. nebulosa and F. catus, metacarpal or metartarsal pads, N = 8, digital pads N = 24. Values with different numbers are significantly different from each other (P ≤ 0.05).

  2. Abbreviations: corn, stratum corneum; noncorn, non-cornified layers.

N. nebulosaMetacarpal pad corn.93.8 ± 1.41
Metatarsal pad corn.97.0 ± 2.81
Metacarpal pad noncorn.24.5 ± 0.22
Metatarsal noncorn.22.5 ± 0.32
Manus digit pad corn.76.5 ± 1.23
Pes digit pad corn.79.4 ± 1.83
Manus digit pad noncorn.17.8 ± 0.64
Pes digit pad noncorn.14.4 ± 0.24
F. catusMetacarpal pad corn.26.1 ± 0.52
Metatarsal pad corn.25.8 ± 0.32
Metacarpal pad noncorn.10.8 ± 0.15
Metatarsal pad noncorn.10.9 ± 0.35
Manus digit pad corn.17.1 ± 0.34
Pes digit pad corn.21.5 ± 0.32
Manus digit pad noncorn.8.5 ± 0.15
Pes digit pad noncorn.10.3 ± 0.25
Figure 10.

The epidermis and dermis of the paw pad of N. nebulosa (×40, H&E). Collagen fiber connective tissue septa can be seen running from the deeper regions of the metatarsal pad and intersecting with fibers running parallel to the epidermis.

Gold-stained sections of digital and metatarsal/metacarpal pads showed an extensive plexus of nerve fibers in the dermis predominantly in the regions immediately deep to the dermal papillary layer (Fig. 11). Individual nerve fibers were seen rising toward the epidermis and connecting with pointed extensions of the stratum basalis of the epidermis in both the digital pads and in the metacarpal/metatarsal pads. Encapsulated sensory endings such as Pacinian or Meissner's corpuscles were not detected in any of the samples examined for either F. catus or the N. nebulosa.

Figure 11.

A photomicrograph of silver stained nerve fibers in the N. nebulosa paw pad connecting with the dermal papillary layer of the stratum basalis of the epidermis (×40).

Thick bands of elastic fibers were present throughout the dermal region, particularly in the deep dermal areas (Fig. 12). Analysis of the digital pads showed a pattern similar to that of the metacarpal/metatarsal pad, however, less adipose tissue and more dense connective tissue was present (Fig. 13).

Figure 12.

A photomicrograph of elastic fibers (black stained fibers Verhoeff's elastin stain) in the dermis of the N. nebulosa manus (×40).

Figure 13.

The digital pad of the N. nebulosa manus (×40, H&E). Adipose tissue compartments can be seen interspersed between thick bundles of collagen fibers. A sweat gland is also present.

Size Comparison of Manus and Pes

The N. nebulosa manus and pes were both significantly wider proportionally than those of F. catus (Table 2). The average length to width ratio for the N. nebulosa manus was 1.055, compared to 0.600 for F. catus (t = 22.671, P < 0.001). For the N. nebulosa pes the ratio was 0.680, and 0.400 for F. catus. (t = 6.160 P < 0.001).

Table 2. Comparison of the manus and pes size ratio in F. catus and N. nebulosa
 Width (cm)Length (cm)W/L Ratio
  1. Analysis using two-sample Student's t test showed significant differences in the W/L ratios between the N. nebulosa and F. catus in both the manus (t = 22.671, P < 0.001) and pes (t = 6.160, P < 0.001).

Manus
 N. nebulosa 14.404.151.06
 N. nebulosa 25.204.951.05
 Mean1.055
 F. catus 11.702.750.62
 F. catus 21.552.700.57
 F. catus 31.752.850.61
 Mean0.600
Pes
 N. nebulosa 13.805.900.64
 N. nebulosa 24.506.200.73
 Mean0.680
 F. catus 11.454.150.35
 F. catus 21.854.400.42
 F. catus 31.904.450.43
 Mean0.400

DISCUSSION

Previous reports have mentioned suspensory ligaments associated with the metacarpal and metatarsal pad in the pes and manus of F. catus (Crouch, 1969; Walker, 1986; Rosenzweig, 1990). In this study, we have observed that F. catus manus has four metacarpal pad suspensory ligaments that project into the connective tissue and skin of the metacarpal pad while the N. nebulosa has five. In the pes the metatarsal pad suspensory ligament morphology is essentially the same for both F. catus and N. nebulosa with the exception that there appear to be four branches to the ligament in the N. nebulosa. In addition, both F. catus and N. nebulosa have numerous small fibers seen both grossly and in histological sections that appear to anchor the deep fascia and tendons to the metatarsal pad. An extensive review of the literature gives only rudimentary descriptions of metacarpal/metatarsal pad suspensory ligament morphology with no suggested function for either. However, it seems likely, given their connection to flexor muscle tendons and the skin, that they could play a role in modifying both the frictional and conformation properties of the manus and pes. This would be important at points where the paw contacts various substrata during walking or climbing. Given the size and arboreal nature of the N. nebulosa it is not surprising that this animal has developed additional structural supportive elements in the manus and pes paw pads when compared to F. catus. The combined effect of these tendinous insertions and the flexibility of the paw pad in felids may allow differential contraction of the pad that would enhance the animal's ability to grip substrates of unpredictable size and shape, such as tree branches, while climbing.

The relative size of the manus and pes as well as the thickness of the pad epidermis was also larger and greater respectively when the N. nebulosa was compared to F. catus. A larger manus and pes would provide better functional support and substrate-gripping capacity for the N. nebulosa when walking on tree limbs. The thicker pad epidermis is likely due to the increased size of the animal but may also be due to increased potential for wear when walking on rough substrates such as tree bark.

The manus of N. nebulosa has been described previously as “spoon-shaped.” This shape has been suggested to be an adaptation for improved ability to grip branches while climbing. An unusual feature of N. nebulosa climbing habits is the ability to engage in suspensory locomotion beneath the branches of trees (Nowell and Jackson, 1996). However, describing the manus as spoon-shaped maybe confusing in that such a shape suggests that the central region of the manus is concave with digits 1 and 5 forming the medial and lateral sides of a rounded hollow. However, a manus of this shape would not allow an animal to grip a tree branch easily, unless it was able to generate sufficient force to create suction, an idea that has never been suggested, and that lacks any anatomical support. Instead, a more accurate description of the manus shape in the N. nebulosa is that it is capable of creating a hook shape by curling the paw to facilitate hanging from branches. This curvature and the anatomical features that maintain it would assist the cat greatly in gripping substrates while climbing. The M. flexor digitorum superficialis tendons to digits 2, 3, and 4 are large and robust, and in a relatively direct line with the axis of contraction of forelimb muscles thus providing more support for these digits when the animals hangs from a branch. By contrast, tendons to digits 1 and 5 are significantly more slender and are more oblique to the forelimb muscle axis and therefore probably play a reduced role in climbing activities. A similar shape is also seen in the hands of many primates and adaptations for suspensory locomotion in this group have been studied extensively (Van Horn, 1972; Turnquist et al., 1983, 1999; Byron and Covert, 2004). Tree sloths also practice suspensory locomotion (Naples, 1986; Nowack, 1999). These groups emphasize the use of the digits that align most closely to the main axis of the forelimb in grasping the substrate, often reducing the size and strength of medial and lateral digits. For example, primates that display hand-over-hand arboreal locomotion, such as gibbons and spider monkeys, often have thumbs that are reduced in size and are not used as a part of this grip (Van Horn, 1972; Turnquist, et al., 1999). Sloths show hands that are even more restricted to grasping curved surfaces by reduction of the digits to three in the bradypodids or even two in choloepids (Nowak, 1999).

This difference in the relative size of the tendons contrasts with the condition in F. catus, where all of the M. flexor digitorum superficialis tendons are of similar size. The increase in the relative size of the tendons in the N. nebulosa paw that are more in direct line with the bellies of the forearm muscles would better support increased tension by these muscles produced while maintaining a hook-shaped paw during suspension beneath a tree limb, and can be considered an adaptation for this specialized kind of locomotion.

Histological analysis of the metacarpal/metatarsal pads in the N. nebulosa and F. catus shows that they are organized into adipose tissue compartments surrounded by dense collagen and elastic fibers. The collagen fibers appear to serve as anchors between the subdermal regions and the epidermis in addition to acting as partitioning agents. Previous studies in the elephant and F. catus indicate that the adipose compartments, combined with the profusion of elastic fibers impart a cushioning effect to the metacarpal and metatarsal pad (Alexander et al., 1986; Ker, 1999; Liebich, 1999; Weissengruber et al., 2006). Adipose tissue compartments have been suggested to act as the equivalent of a fluid-filled cushion where large volume fluid flow is limited during compression by partitioning connective tissue septae in a way that is similar to isolated springs in a bed mattress (Ker, 1999; Weissengruber et al., 2006). This would appear to enhance conformational contouring of the paw pad to variable substrates during walking and climbing as well as providing a cushioning effect. Tensing of the paw pad tendons when the flexor muscles are active in either the manus or pes would appear to stiffen the pad by pulling on the connective tissue elements surrounding the adipose compartments. This may act to stabilize the pad further, and enhance the grip during climbing.

The silver stained slides of the subepidermal region of the paw pads showed a large numbers nerve fibers but the pad was surprisingly devoid of mechanoreceptors such as the Pacinian and Meissner corpuscles that are prevalent in the feet, fingers and digital pads of elephants, humans, primates, certain rodents and marsupials (Weissengruber et al., 2006). This seems unusual given that these structures are useful for sensing variable surfaces to facilitate walking (Weissengruber et al., 2006). While it is possible that these were located in other regions of the pad that were not tested, it seems unlikely because all major regions of the metacarpal/metatarsal pads and digital pads were examined. Of note, however, is the unusual way in which nerves from the dermal region attach to pointed projections of the epidermis (see Fig. 11). It is possible that these represent some undermined sensory modality.

Numerous eccrine sweat glands were present in all histological samples for both the N. nebulosa and F. catus. Previous studies have indicated that sweat from these glands improves the frictional capacity of the paw pad and provides scent markings (Meyer and Bartels, 1989). It is likely the function is similar in F. catus and N. nebulosa.

Acknowledgements

The authors thank Charley Potter and John Ososky and the Smithsonian Institution for facilitating this study.

Ancillary