Muscular anatomy of the forelimb of tiger (Panthera tigris)

Abstract

Dissection reports of large cats (family Felidae) have been published since the late 19th century. These reports generally describe the findings in words, show drawings of the dissection, and usually include some masses of muscles, but often neglect to provide muscle maps showing the precise location of bony origins and insertions. Although these early reports can be highly useful, the absence of visual depictions of muscle attachment sites makes it difficult to compare muscle origins and insertions in living taxa and especially to reconstruct muscle attachments in fossil taxa. Recently, more muscle maps have been published in the primary literature, but those for large cats are still limited. Here, we describe the muscular anatomy of the forelimb of the tiger (Panthera tigris), and compare muscle origins, insertions, and relative muscle masses to other felids to identify differences that may reflect functional adaptations. Our results reiterate the conservative nature of felid anatomy across body sizes and behavioral categories. We find that pantherines have relatively smaller shoulder muscle masses, and relatively larger muscles of the caudal brachium, pronators, and supinators than felines. The muscular anatomy of the tiger shows several modifications that may reflect an adaptation to terrestrial locomotion and a preference for large prey. These include in general a relatively large m. supraspinatus (shoulder flexion), an expanded origin for m. triceps brachii caput longum, and relatively large m. triceps brachii caput laterale (elbow extension), as well as relatively large mm. brachioradialis, abductor digiti I longus, and abductor digiti V. Muscle groups that are well developed in scansorial taxa are not well developed in the tiger, including muscles of the cranial compartment of the brachium and antebrachium, and m. anconeus. Overall, the musculature of the tiger strongly resembles that of the lion (Panthera leo), another large‐bodied terrestrial large‐prey specialist.

We describe the muscular anatomy of the tiger forelimb, and compare muscle origins, insertions, and relative muscle masses to other felids. Our results reiterate the conservative nature of felid anatomy across body sizes and behavioral categories. We find several differences in regional muscle mass between the two felid subfamilies, Pantherinae and Felinae, and identify modifications of tiger musculature that may reflect an adaptation to terrestrial locomotion and a preference for large prey.

1. INTRODUCTION

The forelimbs of the Felidae serve a dual purpose for both locomotion and prey‐killing (Meachen‐Samuels & Van Valkenburgh, 2009) when compared with other carnivoran families such as the Canidae or Ursidae in which the fore‐and hindlimbs serve primarily for locomotion. Therefore, a detailed understanding of the muscular and osteological anatomy of the forelimbs attained through basic anatomical dissection, observation, and measurement of felids is crucial for understanding both of these aspects of felid ecology and behavior. Specifically, knowledge of osteological correlates of muscular anatomy are necessary for reconstruction of behavior in extinct carnivorans through functional morphological analyses (Andersson, 2004; Gonyea, 1978; Heinrich & Biknevicius, 1998; Hildebrand et al., 1985; Iwaniuk et al., 1999; Lanyon & Rubin, 1985; Lewis & Lague, 2011; Panciroli et al., 2017; Polly, 2010; Samuels et al., 2013; Van Valkenburgh, 1987; Walmsley et al., 2012).

Dissections of large carnivores generally, and specifically of felids have been common in the scientific literature since the mid‐19th century (e.g., Barone, 1967; Encinoso et al., 2019; Haughton, 1864; Hubbard et al., 2009; Ross, 1883; Scharlau, 1925; Straus‐Durckheim, 1845; Windle, 1889; Windle & Parsons, 1897, 1898). These types of studies typically describe their dissections in words, accompanied by drawings or photos of the dissection as it proceeded, but usually do not document exactly where muscles attached in the form of a detailed muscle map, with a few exceptions (e.g., Haines, 1950; Hudson et al., 2011; Reighard & Jennings, 1901). Several muscle maps of felid fore‐ and hindlimbs have been published recently, including the forelimb of the ocelot, L. pardalis (Julik et al., 2012), Eurasian lynx, Lynx lynx (Viranta et al., 2016), and snow leopard, Panthera uncia (Smith et al., 2021), and the hindlimb of the clouded leopard, Neofelis nebulosa (Carlon & Hubbard, 2012), margay cat, Leopardus weidii, and Geoffroy's cat L. geoffroyi (Morales et al., 2018), and Eurasian lynx Lynx lynx (Viranta et al., 2021). These types of studies are transformational because they give the exact positions of the muscles in relation to osteological features, which allows direct comparison in muscle attachment locations and areas between species. Not only do these studies facilitate direct comparisons between extant species, but they also allow reconstruction of muscle attachments in fossil taxa, resulting in much more detailed inferences of the behavior of long‐dead species, such as sabertooth cats. Additionally, having these detailed muscle maps available for multiple species in a family (such as Felidae) makes it possible to identify muscles that show conservative attachments within that family and those that are more variable, thus identifying muscles that may be good indicators of functional differences between closely related species.

Although the muscular anatomy of large cats, particularly the lion (Panthera leo), is relatively well documented (Barone, 1967; Cuvier & Laurillard, 1850; Ellenberger et al., 1949; Haughton, 1864; Sánchez et al., 2019; Scharlau, 1925), the snow leopard (Panthera uncia) is the only big cat for which a detailed and comprehensive muscle map is published (Smith et al., 2021). This lack of information impedes our understanding of big cat functional morphology in a variety of ways. First, it limits our knowledge about allometric differences in muscle attachment areas between large and small felid species (e.g., do large felids have larger muscle attachment areas than small felids? Are there differences in the position of muscle attachments in large cats vs. small cats?). Additionally, it precludes inferences about whether similar functional adaptations are reflected in similar muscular anatomy in small vs. large cats (e.g., does an increased range of pronation/supination ability in large vs. small cats correlate with similar muscular modifications?). Finally, it limits the accuracy of muscular and functional reconstructions of large fossil felids. In this paper we wish to address this fundamental gap by producing detailed origin and insertion maps along with a description of the muscular anatomy of the forelimb of the tiger. These maps can serve as a direct comparison to those of smaller cats (Julik et al., 2012; Reighard & Jennings, 1901; Viranta et al., 2016), and as a more accurate template for reconstructing muscle attachments of extinct large pantherine cats, such as the American lion (Panthera atrox), and more distantly related sabertooth cat species (subfamily Machairodontinae). We further compare the muscular anatomy, origins, insertions, and muscle mass distribution of the tiger with other felids to identify functional adaptations in forelimb musculature.

2. MATERIALS AND METHODS

The forelimbs in the current study belonged to an 18‐year‐old female Amur tiger (Panthera tigris altaica) from Blank Park Zoo in Des Moines, IA. Although the tiger was of an advanced age, the joints of the forelimbs showed no signs of degeneration, osteoarthritis, or other pathology. The limbs were detached from the thorax by zoo personnel following necropsy, and subsequently frozen and dissected at Des Moines University. The dissection sessions were video recorded without audio. At each stage of dissection, muscles were identified and digitally photographed. In the dissection of the left forelimb, the origin and insertion of each muscle were drawn directly on the bones when possible and photographed for reference. For the right forelimb, preliminary muscle maps were created by drawing muscle attachments on print‐out screen captures of 3D models the forelimb bones of a Bengal Tiger, downloaded from MorphoSource.org (Museum of Vertebrate Zoology specimen 189,634). The final muscle maps were created from these preliminary maps, video recordings, photographs, and the marks remaining on the bones. Unless otherwise noted, the attachment and muscle patterns were the same for the right and left limbs.

Muscle mass (in grams) for each muscle was recorded using an Ohaus Scout Pro (model SP401) digital scale and is provided in Table 1. We did not weigh muscles that were damaged in the necropsy nor those that were fused to other muscles such that they could not be separated reliably (e.g., m. teres major). Tendons were removed prior to weighing. We averaged the muscle mass from the right and left sides and used those values to calculate total and regional muscle mass of the forelimb as well as proportional muscle mass for each region and muscle (Table 1). For statistical analysis, we calculated the proportional mass of muscles grouped in two separate ways: by region, creating five regional variables, and by function, creating ten functional variables, which facilitated comparison with more felid taxa (Cuff et al., 2016; Hudson et al., 2011; Julik et al., 2012; Smith et al., 2021; Souza Junior et al., 2021; Souza Junior et al., 2018; Viranta et al., 2016). The functional groups follow the methods of Souza Junior et al. (2018); the muscles within functional and regional groups are listed in Table 2. We divided the felid taxa into locomotor, subfamilial, and prey preference categories following previous studies (Table 3) (Meachen‐Samuels & Van Valkenburgh, 2009; Smith et al., 2021). Data from multiple studies for the same species were combined using weighted averages when appropriate. All variables except for body mass and carpal extensors (ExC) were normally distributed; natural‐log transformation of body mass resulted in a normal distribution for this variable; however, ExC remained non‐normal after log‐transformation. Because of this and the small sample size, we used a non‐parametric Kruskal–Wallis test to identify statistically significant differences in regional and functional variables between behavioral and phylogenetic groups. We used Spearman's rho to test all variables for correlations with the natural log of body mass. When available, we used body mass data from the same publication as the muscle mass data, and used species‐average body mass values when necessary (Table 3). Species‐average body mass was taken from the PanTHERIA database (Jones et al., 2009) for Panthera leo and Leopardus pardalis muscle data provided in Julik et al. (2012) and the P. tigris data presented here. Body mass data for L. colocolo and L. geoffroyi were taken from Anile and Devillard (2020).

TABLE 1.

Right, left, and average muscle mass (g) and proportional muscle mass (%) in our tiger specimen

Muscle	Right	Left	Average	%
Intrinsic shoulder
M. deltoideus	221.20	223.10	222.15	11.74
Pars scapularis	110.90	117.30	114.10	6.03
Pars acromialis	110.30	105.80	108.05	5.71
M. supraspinatus	637.50	655.00	646.25	34.15
M. infraspinatus	480.40	453.00	466.70	24.66
M. teres minor	25.80	23.30	24.55	1.30
M. subscapularis	510.90	554.30	532.60	28.15
Total intrinsic shoulder	1875.80	1908.70	1892.25	35.00
Caudal brachium
M. triceps brachii	1482.10	1527.30	1504.70	91.12
Caput longum	890.50	882.60	886.55	53.69
Caput laterale	367.00	426.80	396.90	24.04
Caput mediale	100.50	106.20	103.35	6.26
Caput accessorium	99.20	88.20	93.70	5.67
Caput mediale accessorium	24.90	23.50	24.20	1.47
M. anconeus	63.10	60.30	61.70	3.74
M. tensor fascia antebrachii	–	84.90	84.90	5.14
Total caudal brachium	–	1672.50	1651.30	30.55
Cranial brachium
M. articularis humeri	6.20	4.50	5.35	1.21
M. biceps brachii	294.60	287.70	291.15	65.78
M. brachialis	147.80	144.40	146.10	33.01
Total cranial brachium	448.60	436.60	442.60	8.19
Caudal antebrachium
M. brachioradialis	68.10	73.70	70.90	14.15
M. extensor carpi radialis (longus + brevis)	160.10	167.00	163.55	32.63
Longus	78.20	–	78.20	15.60
Brevis	81.90	–	81.90	16.34
M. extensor carpi ulnaris	68.10	77.80	72.95	14.56
M. extensor digitorum communis	68.20	61.80	65.00	12.97
M. extensor digitorum lateralis	26.70	23.40	25.05	5.00
M. extensor digiti I et II	14.20	14.00	14.10	2.81
M. abductor digiti I longus	67.10	63.30	65.20	13.01
M. supinator	26.80	22.00	24.40	4.87
Total caudal antebrachium	499.30	503.00	501.15	9.27
Cranial antebrachium
M. pronator teres	82.80	85.50	84.15	10.38
M. flexor carpi radialis	44.00	43.60	43.80	5.40
M. palmaris longus	140.80	128.90	134.85	16.63
M. flexor carpi ulnaris	154.10	137.30	145.70	17.97
Caput humerale	48.00	43.40	45.70	5.64
Caput ulnare	106.10	93.90	100.00	12.33
M. flexor digitorum superficialis	9.50	8.00	8.75	1.09
Medial belly	2.30	2.00	2.15	0.27
Intermediate belly	5.40	4.10	4.75	0.59
Lateral belly	1.80	1.90	1.85	0.23
M. flexor digitorum profundus	349.20	343.30	346.25	42.70
Caput humerale mediale	95.30	50.60	72.95	9.00
Caput humerale laterale	91.40	91.50	91.45	11.28
Caput humerale profundus	39.50	44.40	41.95	5.17
Caput ulnare	51.00	89.20	70.10	8.64
Caput radiale et ulnare	72.00	67.60	69.80	8.61
M. flexor digitorum brevis manus	7.70	7.90	7.80	0.96
Medial superficial belly	3.50	2.20	2.85	0.35
Lateral superficial belly	2.20	3.30	2.75	0.34
Deep belly	2.00	2.40	2.20	0.27
M. pronator quadratus	38.90	40.30	39.60	4.88
Total cranial antebrachium	827.00	794.80	810.90	15.00
Manus
M. abductor digiti V	12.80	11.10	11.95	11.08
M. abductor et opponens digiti I	0.20	–	0.20	0.19
Mm. lumbricales	9.50	9.70	9.60	8.90
Digit II	1.60	1.60	1.60	1.48
Digit III	2.30	2.60	2.45	2.27
Digit IV	2.00	3.50	2.75	2.55
Digit V	3.60	2.00	2.80	2.60
Mm. adductores digitorum	–	10.30	10.30	9.55
Digit I	2.60	2.20	2.40	2.23
Digit II	–	3.20	3.20	2.97
Digit V	4.90	4.90	4.90	4.54
Mm. flexores breves profundi	79.20	72.40	75.80	70.28
Digit I	1.20	1.50	1.35	1.25
Digit II	19.30	15.80	17.55	16.27
Digit III	22.90	21.70	22.30	20.68
Digit IV	20.10	19.70	19.90	18.45
Digit V	15.70	13.70	14.70	13.63
Total manus	–	–	107.85	1.99
Total muscle mass			5406.05

Note: Proportional mass is calculated as average mass of the muscle × 100/total regional muscle mass, or average total regional mass × 100/total muscle mass.

TABLE 2.

Composition of regional and functional groups used in this study

Functional groups	Regional groups
Shoulder extensors (ExS)	Shoulder (S)
M. supraspinatus	M. deltoideus
M. articularis humeri	M. supraspinatus
Shoulder flexors (FlS)	M. infraspinatus
M. deltoideus	M. subscapularis
M. infraspinatus	M. teres minor
M. teres minor	M. teres major
M. teres major	Caudal brachium (CaB)
Elbow extensors (ExE)	M. triceps brachii
M. triceps brachii	M. anconeus
M. anconeus	M. tensor fascia antebrachia
M. tensor fascia antebrachii	Cranial brachium (CrB)
Elbow flexors (FlE)	M. articularis humeri
M. biceps brachii	M. biceps brachii
M. brachialis	M. brachialis
Carpus extensors (ExC)	Caudal antebrachium (CaAb)
M. extensor carpi radialis	M. brachioradialis
Carpus flexors (FlC)	M. extensor carpi radialis
M. extensor carpi ulnaris	M. extensor carpi ulnaris
M. flexor carpi radialis	M. extensor digitorum communis
M. flexor carpi ulnaris	M. extensor digitorum lateralis
Digital extensors (ExD)	M. extensor digiti I et II
M. extensor digitorum communis	M. abductor digit I longus
M. extensor digitorum lateralis	M. supinator
M. extensor digiti I et II	Cranial antebrachium (CrAb)
M. abductor digiti I longus	M. pronator teres
Digital flexors (FlD)	M. flexor carpi radialis
M. palmaris longus	M. palmaris longus
M. flexor digitorum profundus	M. flexor carpi ulnaris
Supinators (Sup)	M. flexor digitorum profundus
M. supinator	M. pronator quadratus
M. brachioradialis
Pronators (Pron)
M. pronator teres
M. pronator quadratus

Note: Functional groups follow Souza Junior et al. (2018).

TABLE 3.

Taxa included in quantitative analysis in this study

Taxon	N	Subfamily	Locomotion a	Prey size a	Body mass (g)	Analysis
Acinonyx jubatus b	8	Felinae	Terrestrial	Large	33,100	FG, RG
Caracal caracal c	1	Felinae	Scansorial	Mixed	6600	FG, RG
Felis nigripes c	1	Felinae	Terrestrial	Small	1100	FG, RG
Felis sylvestris c	1	Felinae	Scansorial	Small	2660	FG, RG
Leopardus colocolo d	2	Felinae	Scansorial	Small	3950 e	FG
Leopardus geoffroyi d	8	Felinae	Terrestrial	Small	4650 e	FG
Leopardus pardalis c	1	Felinae	Scansorial	Mixed	9600	FG, RG
Leopardus pardalis f	1	Felinae	Scansorial	Mixed	11,880 g	FG, RG
Lynx lynx h	4	Felinae	Scansorial	Large	15,575	FG, RG
Panthera onca c	1	Pantherinae	Scansorial	Large	44,000	FG, RG
Panthera leo c	1	Pantherinae	Terrestrial	Large	133,000	FG, RG
Panthera leo f	2	Pantherinae	Terrestrial	Large	158,624 g	FG, RG
Panthera tigris i	1	Pantherinae	Terrestrial	Large	161,915 g	FG, RG
Panthera tigris c	1	Pantherinae	Terrestrial	Large	86,000	FG, RG
Panthera uncia c	1	Pantherinae	Scansorial	Large	36,000	FG, RG
Panthera uncia j	2	Pantherinae	Scansorial	Large	34,750	FG
Puma yaguarondi d	3	Felinae	Scansorial	Small	6875	FG

Note: N = number of individuals, FG = included in analysis of functional muscle groups, RG = included in analysis of regional muscle groups. Sources designated with superscripts.

^a Meachen‐Samules and VanValkenburgh (2009).

^b Hudson et al. (2011).

^c Cuff et al. (2016).

^d Souza Junior et al. (2021).

^e Anile and Devillard (2020).

^f Julik et al. (2012).

^g Jones et al. (2009).

^h Viranta et al. (2016).

ⁱ This study.

^j Smith et al. (2021).

3. RESULTS

3.1. Quantitative analyses

3.1.1. Regional groups

Two out of the five regional variables were significantly correlated with body mass: shoulder (S) and caudal brachium (CaB). S was negatively correlated with body mass (rho = −0.685; p = 0.035), whereas CaB (rho = 0.903; p = 0.001) was positively correlated. Two regional groups showed significant differences between subfamilies, CaB and cranial brachium (CrB) (Table 4), with pantherines having more massive muscles of the caudal brachium (p = 0.019) and less massive muscles of the cranial brachium (p = 0.019).

TABLE 4.

Summary statistics for regional muscle groups used in this study broken down by category

		Regional groups
		S			CaB			CrB			CaAb			CrAb
Group	N	Mean	SD	Range	Mean	SD	Range	Mean	SD	Range	Mean	SD	Range	Mean	SD	Range
Subfamily
Pantherinae	4	38.30	2.53	35.02–41.12	30.59*	1.11	29.60–32.17	7.93*	0.51	7.30–8.49	9.09	0.85	8.31–9.85	14.09	1.02	12.80–15.20
Felinae	6	43.24	3.69	38.74–47.62	27.92*	1.58	25.88–29.94	8.83*	0.37	8.33–9.39	7.73	1.71	4.66–9.41	12.28	1.91	9.24–14.31
Locomotion
Terrestrial	4	42.26	6.33	35.02–47.62	29.59	2.65	25.88–32.17	8.31	0.39	7.76–8.64	7.53	2.2	4.66–9.85	12.31	2.71	9.24–15.20
Scansorial	6	40.60	1.94	38.16–43.25	28.59	1.39	26.20–30.23	8.58	0.75	7.30–9.39	8.77	0.8	7.78–9.81	13.47	0.90	12.59–14.49
Prey preference
Large	6	39.91	4.21	35.02–47.52	30.12	1.22	28.44–32.17	8.24	0.64	7.30–9.10	8.41	1.96	4.66–9.85	13.32	2.15	9.24–15.20
Mixed	2	42.65	0.84	42.06–43.25	27.19	1.40	26.20–28.17	9.11	0.40	8.82–9.39	8.46	0.96	7.78–9.14	12.60	0.00	12.59–12.60
Small	2	43.95	5.19	40.29–47.62	27.38	2.11	25.88–28.87	8.51	0.24	8.33–8.68	7.69	0.66	7.22–8.15	12.47	2.17	10.94–14.01

Note: Numbers are percent of the total muscle mass; asterisks indicate a significant difference in the mean (p < 0.05). N = sample size, SD = standard deviation. Abbreviations are found in Table 2.

3.1.2. Functional groups

Results from comparisons of functional groups are similar to those for regional groups. Three out of ten functional variables were significantly correlated with body mass: elbow extensors (ExE; rho = 0.764; p = 0.004), supinators (Sup; rho = 0.560; p = 0.049), and pronators (Pron; rho = 0.579; p = 0.021); all were positively correlated with body mass. ExE, Sup, Pron, shoulder extensors (ExS), and shoulder flexors (FlS) showed significant differences between subfamilies (Table 5). No functional variables showed significant differences among prey or locomotor categories. Pantherines have significantly smaller ExS (p = 0.031) and FlS (p = 0.045), and larger ExE (p = 0.005), Sup (p = 0.031), and Pron (p = 0.021) than felines.

TABLE 5.

Summary statistics for functional muscle groups used in this study broken down by category

		Functional groups
		ExS			FlS			ExE			FlE			ExC			FlC			ExD			FlD			Sup			Pron
Group	N	Mean	SD	Range	Mean	SD	Range	Mean	SD	Range	Mean	SD	Range	Mean	SD	Range	Mean	SD	Range	Mean	SD	Range	Mean	SD	Range	Mean	SD	Range	Mean	SD	Range
Subfamily
Pantherinae	4	12.43*	0.81	11.39–1336	19.95*	1.49	18.54–22.02	33.60*	0.24	33.33–33.85	8.63	0.84	7.59–9.35	2.96	0.71	2.18–3.83	6.06	2.08	4.27–8.88	3.81	0.77	2.86–4.60	7.91	1.3	6.16–9.15	2.01*	0.43	1.44–2.47	2.63*	0.23	2.42–2.96
Felinae	9	15.31*	2.16	11.26–17.79	21.85*	1.95	20.04–25.39	31.38*	1.31	29.55–33.21	8.67	0.92	6.94–9.78	3.11	0.99	1.02–3.93	4.81	0.90	3.06–5.77	3.51	0.71	1.95–4.09	7.89	0.93	6.61–9.70	1.32*	0.43	0.92–1.94	2.14*	0.46	1.19–2.91
Locomotion
Terrestrial	5	14.43	2.38	11.39–17.79	21.55	3.15	18.54–25.39	32.36	1.74	29.55–33.85	8.71	0.72	7.66–9.35	2.54	1.12	1.02–3.93	5.00	1.24	3.06–6.37	3.61	1.00	1.95–4.60	7.90	1.09	6.61–9.15	1.65	0.52	0.98–2.09	2.25	0.66	1.19–2.96
Scansorial	8	14.42	2.38	11.26–17.42	21.09	0.99	19.94–23.23	31.88	1.45	29.73–33.74	8.62	0.98	6.94–9.78	3.39	0.55	2.18–3.83	5.32	1.57	4.04–8.88	3.61	0.54	2.86–4.23	7.89	1.02	6.16–9.70	1.46	0.55	0.92–2.47	2.32	0.32	1.92–2.91
Prey preference
Large	6	13.13	2.42	11.26–17.79	21.08	2.22	18.54–24.49	33.06	1.08	30.95–33.85	8.94	0.82	7.59–9.78	2.78	1.07	1.02–3.83	5.44	2.01	3.06–8.88	3.52	0.97	1.95–4.60	8.05	1.36	6.16–9.70	1.69	0.59	0.93–2.47	2.31	0.61	1.19–2.96
Mixed	2	15.49	1.18	14.66–16.32	21.78	0.64	21.33–22.23	30.45	1.02	29.73–31.17	8.71	0.11	8.63–8.79	3.31	0.38	3.04–3.57	4.90	1.23	4.04–5.77	3.40	0.73	2.89–3.92	7.89	0.46	7.56–8.21	1.43	0.73	0.92–1.94	2.64	0.38	2.37–2.91
Small	5	15.55	1.8	13.41–17.42	21.29	2.32	20.04–25.39	31.51	1.41	29.55–33.21	8.29	1.03	6.94–9.56	3.32	0.8	1.93–3.93	5.02	0.58	4.31–5.70	3.79	0.33	3.26–4.09	7.72	0.76	6.61–8.59	1.38	0.42	0.93–1.93	2.14	0.18	1.92–2.40

Note: Numbers are percent of the total muscle mass; asterisks indicate a significant difference in the mean (p < 0.05). N = sample size, SD = standard deviation. Abbreviations for muscle group names are found in Table 1.

3.2. Extrinsic muscles of the shoulder

The origins, and in some cases the bulk, of these muscles were removed when the upper limbs were detached from the thorax following the necropsy; consequently, the identity of some muscles is ambiguous. Here we describe the insertion of these muscles to the best of our ability.

3.2.1. M. trapezius

The origin of m. trapezius as well as most of the muscle bulk was removed during the necropsy. Nevertheless, two parts were distinguishable: the pars cervicalis and pars thoracica. M. trapezius pars cervicalis has a more extensive insertion, beginning just caudal to the insertion of m. omotransversarius and extending the length of the scapular spine (Figure 1). In addition to this bony insertion, m. trapezius pars cervicalis also has an extensive attachment to the aponeurosis overlying m. infraspinatus. M. trapezius pars thoracica has muscle fibers oriented perpendicular to the pars cervicalis. It inserts dorsal to the pars cervicalis, superficial to m. rhomboideus, and somewhat deep to m. trapezius pars cervicalis.

FIGURE 1.

Scapular muscle maps for Panthera tigris (left side). (a) Medial surface; (b) lateral surface; (c) caudal border. Light gray = origins, dark gray = insertions. Muscle abbreviations: Ah, articularis humeri; Bb, biceps brachii, Dac, deltoideus pars acromialis; Dsc, deltoideus pars scapularis; Isp, infraspinatus; Otr, omotransversarius; Rcap, rhomboideus capitis; Rcerv, rhomboideus cervicis; Rthor, rhomboideus thoracis; Ssc, subscapularis; Ssp, supraspinatus; Sv, serratus ventralis; Tblong, triceps brachii caput longum; Tcerv, trapezius pars cervicis; Tthor, trapezius pars thoracica; Tmj, teres major; Tmn, teres minor. [Planned for page width]

3.2.2. M. omotransversarius

This muscle forms a narrow strip ventral to m. trapezius pars cervicalis. Although it is contained within the same investing fascia, it is easily separable from m. trapezius and has a distinct fleshy insertion onto the suprahamate process of the acromion near the origin of m. deltoideus pars scapularis.

3.2.3. M. latissimus dorsi

Musculus latissimus dorsi splits into portions that run on either side of the belly of m. biceps brachii to insert onto the humerus in two places. The superficial division of m. latissimus dorsi joins the caudal belly of m. pectoralis profundus to insert onto the medial edge of the deltopectoral crest. The deep portion of this muscle joins m. teres major to insert onto the teres major tuberosity (tuberositas teres major) of the medial humerus via a robust tendon (Figure 2).

FIGURE 2.

Humeral muscle maps for Panthera tigris (left side). (a) Proximal view; (b) caudal view; (c) distal view; (d) medial view; (e) lateral view; (f) cranial view. Light gray = origins, dark gray = insertions, black and white striped areas = ligament attachments. Muscle abbreviations: Ah, articularis humeri; An, anconeus; Br, brachialis; Brad, brachioradialis; Ecrb, extensor carpi radialis brevis; Ecrl, extensor carpi radialis longus; Ecu, extensor carpi ulnaris; Edc, extensor digitorum communis; Edl, extensor carpi lateralis; Dt, deltoideus; Fcr, flexor carpi radialis; Fcuh, flexor carpi ulnaris caput humerale; Fdphl, flexor digitorum profundus caput humerale laterale; Fdphm, flexor digitorum profundus caput humerale mediale; Fdphp, flexor digitorum profundus caput humerale profundus; Isp, infraspinatus; Ld, latissimus dorsi; Pl, palmaris longus; Ppca, pectoralis profundus caudal head; Ppcr, pectoralis profundus cranial head; Ps, pectoralis superficialis; Pt, pronator teres; Rcl, radial collateral ligament Ssc, subscapularis; Ssp, supraspinatus; Tba, triceps brachii caput accessorium; Tblat, triceps brachii caput laterale; Tbm, triceps brachii caput mediale; Tbma, triceps brachii caput mediale accessorium; Thl, transverse humeral ligament; Tmj, teres major; Tmn, teres minor; Ucl, Ulnar collateral ligament. [Planned for page width]

3.2.4. M. rhomboideus

This muscle typically consists of three parts: mm. rhomboideus capitis, cervicis, and thoracis. These muscles were extensively damaged in the necropsy, with only a very small portion near their insertion present. We were able to discern two insertions along the vertebral border of the scapula for m. rhomboideus, one at the cranial angle just dorsal to the termination of the scapular spine and the other along the vertebral border extending to the caudal angle. We suspect that the cranial insertion is for the fused mm. rhomboideus capitis and cervicis, as is common in felids (Smith et al., 2021; Viranta et al., 2016), whereas the longer insertion is for m. rhomboideus thoracis. The majority of muscle fibers inserted on the lateral surface of the scapula, but several fibers also inserted onto the costal surface.

3.2.5. M. serratus ventralis

The only portion of this muscle remaining following the necropsy was the insertion, so we were not able to separate it into costal versus vertebral segments. The bulk of this muscle inserts onto the cranial angle of the costal surface of the scapula but some fibers extend caudally toward the caudal angle.

3.2.6. M. pectoralis superficialis

Because the forelimbs were detached from the thorax following the necropsy, distinct origins could not be used to distinguish m. pectoralis superficialis from profundus and we had to rely on the points of insertion. M. pectoralis superficialis appears to consist of two parts: the larger proximal part inserts via fleshy fibers onto the lateral deltopectoral crest of the humerus and distolateral portion of the greater tubercle (Figure 2), and the distal part fuses with mm. cleidobrachialis and brachialis and inserts with them onto the medial surface of the proximal ulna just caudal to the trochlear notch (Figure 3).

FIGURE 3.

Radial and ulnar muscle maps for Panthera tigiris (left side). (a) Proximal view; (b) cranial view; (c) lateral view; (d) medial view; (e) caudal view. Light gray = origins, dark gray = insertions, black and white striped areas = ligament attachments. Muscle abbreviations: Ab1long, abductor digiti I longus; An, anconeus; Bb, biceps brachii; Br, brachialis; Brad, brachioradialis; Cbr, cleidobrachialis; Dal, deep antebrachial ligament; Ed12, extensor digiti I et II; Fcuu, flexor carpi ulnaris caput ulnare; Fdpru, flexor digitorum profundus caput radiale et ulnare; Fdpu, flexor digitorum profundus caput ulnare; Iom, interosseous membrane; Pq, pronator quadratus; Ps, pectoralis superficialis; Pt, pronator teres; Rcl, cadial collateral ligmanet; Su, supinator; Tba, triceps brachii caput accessorium; Tblat, triceps brachii caput laterale; Tblong, triceps brachii caput longum; Tbm, triceps brachii caput mediale; Tbma, triceps brachii caput mediale accessorium; Ucl, ulnar collateral ligament. [Planned for page width]

3.2.7. M. pectoralis profundus

Musculus pectoralis profundus was divisible into two bellies with separate insertions, the cranial and caudal bellies. Because the origin of m. pectoralis profundus was removed, the presence of an abdominal belly could not be assessed; however, it may have been fused with the caudal belly as in the ocelot and red panda (Fisher et al., 2009; Julik et al., 2012). The cranial belly inserts onto the medial aspect of the greater tubercle of the humerus via a short, distinct tendon inferior to the insertion of m. supraspinatus (Figure 2). On the right side, this muscle was fused with m. supraspinatus at its insertion. The caudal belly fuses with fibers of m. latissimus dorsi to form a thin, flat tendon that inserts onto the medial edge of the deltopectoral crest of the humerus and the distomedial edge of the greater tubercle.

3.2.8. M. brachiocephalicus

Three portions of this muscle were discernable: mm. cleidomastoideus, cleidocephalicus, and cleidobrachialis. The origins of mm. cleidomastoideus and cleidocephalicus were not present but m. cleidomastoideus appears to be the smallest member of this muscular complex. M. cleidomastoideus is located deep to the cleidocephalicus and inserts onto the caudal two‐thirds of the clavicle deep to the insertion of m. cleidocephalicus. M. cleidocephalicus inserts onto the caudal two‐thirds of the clavicle and to the intersectio clavicularis, a fibrous raphe caudal to the clavicle. The muscular fibers of m. cleidocephalicus continue distally from their insertion onto the clavicle and intersectio clavicularis to form the cleidobrachialis muscle, which continues into the brachium and is fused to m. pectoralis superficialis for approximately the distal two‐thirds of its length. The combined fibers of mm. cleidobrachialis and pectoralis profundus join the tendon of m. brachialis and insert onto the proximal ulna (Figure 3).

3.3. Intrinsic muscles of the shoulder

We were unable to sufficiently separate m. teres major from m. latissimus dorsi to get an accurate weight but the remaining muscles of the intrinsic shoulder comprise 35.0% of the overall mass of the forelimb, more than any other region. Within this region, m. supraspinatus is the largest, making up 34.15% of the total muscle mass of intrinsic shoulder muscles. Mm. infraspinatus and subscapularis each makes up approximately one quarter of shoulder muscle mass (24.66% and 28.15%, respectively), followed by m. deltoideus (11.74%), and m. teres minor (1.30%).

3.3.1. M. deltoideus

M. deltoideus consists of two distinct heads, the pars acromialis and pars scapularis, that are roughly equal in size and originate separately. M. deltoideus pars scapularis was the larger of the two heads in both limbs; however, the difference between the two was larger in the left limb than in the right (Table S1). M. deltoideus pars acromialis originates from the hamate process and m. deltoideus pars scapularis from the suprahamate process of the acromion and the fascia overlying m. infraspinatus (Figure 1). The two parts fuse distally and insert in common via tendinous and fleshy fibers onto the deltoid tuberosity of the humerus just distal and caudal to the insertion of m. pectoralis superficialis (Figure 2). Some fibers of m. deltoideus pars scapularis insert onto the investing fascia of m. triceps brachii caput laterale.

3.3.2. M. supraspinatus

M. supraspinatus is the largest muscle of the shoulder, making up approximately one‐third of the total muscle mass of the intrinsic shoulder muscles. This muscle originates from the supraspinous fossa, cranial border of the scapula, and cranial aspect of the scapular spine (Figure 1). At the cranial border of the scapula, it partially fuses with m. subscapularis. It inserts onto the proximal aspect of the greater tubercle of the humerus superior to the insertion of m. pectoralis profundus cranial belly (Figure 2).

3.3.3. M. infraspinatus

M. infraspinatus originates from the infraspinous fossa, caudal scapular spine, and tendon of origin m. triceps brachii caput longum (Figure 1). It inserts via a distinct tendon onto a raised, circular ridge on the lateral aspect of the greater tubercle of the humerus (Figure 2).

3.3.4. M. teres minor

M. teres minor is a small muscle that originates primarily from the tendon of m. triceps brachii caput longum as well as from the scapular neck just caudal to the glenoid fossa (Figure 1). It inserts via fleshy fibers onto the distal margin of the greater tubercle of the humerus (Figure 2) and partially fuses to the joint capsule.

3.3.5. M. teres major

M. teres major originates from the caudal angle and caudal border of the scapula, where it fuses with m. subscapularis (Figure 1). The cranial extent of the bony origin of m. teres major is well‐delineated by a strong ridge. This muscle combines with the deep portion of m. latissimus dorsi to insert onto the teres major tuberosity of the medial humerus (Figure 2).

3.3.6. M. subscapularis

Musculus subscapularis originates from the subscapular fossa of the scapula (Figure 1). This muscle fuses cranially with m. supraspinatus, and caudally with m. teres major. It inserts via a thick tendon onto the lesser tubercle of the humerus (Figure 2). Multiple pennations were present in both limbs but the exact number was not recorded.

3.4. Caudal compartment of the brachium

The caudal brachium is the second largest compartment of the forelimb by mass, making up 30.55% of the total muscle mass of the forelimb. M. triceps brachii caput longum comprises 53.69% of the muscle mass of the caudal brachium, followed by the m. triceps brachii caput laterale at 24.04%. Each of the other heads of m. triceps brachii, and mm. anconeus and tensor fascia antebrachii represents less than 10% of the caudal brachium, with m. triceps brachii caput mediale accessorium being the smallest (1.47%).

3.4.1. M. triceps brachii

We identified five heads of the triceps brachii: capita longum, laterale, mediale, accessorium, and mediale accessorium. M. triceps brachii caput longum is the largest head, making up over half of the total mass of m. triceps brachii. It originates from the caudal border of the scapula via a thin tendon extending from the glenoid fossa onto the crest formed by the origin of m. teres major, overlapping with the origin of that muscle (Figure 1c). It inserts onto the caudal olecranon process of the ulna (Figure 3a). M. triceps brachii caput laterale originates from the proximal half of a crest distal to the greater tubercle of the humerus (Figure 2). Its fibers course distally to join with m. triceps brachii caput longum, but it maintains a separate insertion onto the lateral olecranon process just posterior to the insertion of m. anconeus (Figure 3e). M. triceps brachii caput mediale takes origin from the cranial humeral shaft medial to the origin of mm. triceps caput accessorium and brachioradialis (Figure 2). This muscle fuses with the m. triceps brachii caput accessorium proximal to the elbow but maintains a separate, fleshy insertion on the medial lip of the cranial olecranon process cranial to the insertion for the m. triceps brachii caput mediale accessorium (Figure 3).

M. triceps brachii caput accessorium originates from the proximal third of the caudal surface of the humeral shaft including the neck distal to the humeral head (Figure 2). The muscle belly is fused with that of m. triceps brachii caput mediale in the distal brachium; fibers of m. triceps brachii caput accessorium form a flat tendon that inserts onto the olecranon process cranial to the insertion of m. triceps brachii caput longum (Figure 3a). M. triceps brachii caput mediale accessorium is the smallest head of m. triceps brachii. It originates from the caudal surface of the medial epicondyle of the humerus, including the bony strut medial to the entepicondylar foramen (Figure 2) and inserts onto the medial olecranon process caudal to m. triceps brachii caput mediale (Figure 3).

3.4.2. M. anconeus

M. anconeus is a small triangular muscle deep to the distal portion of m. triceps brachii caput mediale. It originates on the caudal aspect of the distal humeral shaft, and to the proximal lateral supracondylar ridge distal to the origin of m. triceps brachii caput mediale and medial to the origin of m. extensor digitorum communis (Figure 2). It inserts onto the lateral surface of the olecranon process (Figure 3).

3.4.3. M. tensor fascia antebrachii

This muscle originates from the superficial surface of the latissimus dorsi and inserts onto the antebrachial fascia of the medial and caudal forearm via an aponeurosis overlying the distal muscle belly of m. triceps brachii caput longum.

3.5. Cranial compartment of the brachium

This compartment comprises 8.19% of the total muscle mass of the upper limb. M. biceps brachii represents 65.78% of the muscle mass of the compartment, whereas m. brachialis makes up 33.01%, followed by m. articularis humeri with 1.21%.

3.5.1. M. articularis humeri

This very small muscle originates from the coracoid process of the scapula via a slender tendon (Figure 1) and inserts via fleshy fibers on the medial aspect of the proximal humerus distal to the lesser tubercle (Figure 2).

3.5.2. M. biceps brachii

M. biceps brachii consists of one head, equivalent to the caput longum and is the largest muscle of thick, cord‐like tendon that travels through the intertubercular sulcus deep to the transverse humeral ligament before expanding into a large, fusiform muscle belly. Distally, it converges into a short tendon to insert onto the radial tuberosity (Figure 3). The bicipital aponeurosis extends from the tendon to blend with the antebrachial fascia overlying the common flexor mass.

3.5.3. M. brachialis

M. brachialis originates from the lateral aspect of the humeral shaft, covering nearly the entire length of the humerus from just inferior to the insertion for m. teres minor, extending distally past the lateral supracondylar ridge (Figure 2). Its origin is intimately related to those of mm. triceps brachii caput laterale and brachioradialis. The muscle belly wraps around the shaft of the humerus to form a tendon that passes cranial to the elbow joint. Proximal to the elbow, the muscle belly of m. brachialis joins the combined tendon of mm. cleidobrachialis + pectoralis superficialis that inserts onto a rough area on the medial aspect of the proximal ulna, just caudal to the coronoid process and distal to the attachment of the ulnar collateral ligament (Figure 3).

3.6. Caudal compartment of the antebrachium

The caudal antebrachium makes up 9.27% of the total forelimb muscle mass. Together, mm. extensor carpi radialis longus (15.60%) and brevis (16.34%) represent 32.63% of muscle mass (the fact that the individual percentages do not add up to 32.63% is an artefact of averaging, as the two muscles were measured separately on the right side but together on the left). Mm. brachioradialis and extensor carpi ulnaris each contributes 14.15% and 14.56%, respectively. Mm. extensor digitorum communis and abductor digiti I longus each represents 12.97% and 13.01%, respectively, and m. extensor digitorum lateralis makes up 5.00% of the mass of the caudal antebrachium. M. extensor digiti I et II is the smallest muscle of this compartment at 2.81%.

3.6.1. M. brachioradialis

This muscle originates from the middle third of the caudal aspect of the humeral shaft between the origins of mm. brachialis and triceps brachii caput mediale and proximal to the origin of m. extensor carpi radialis longus (Figure 2). This broad, flat muscle courses disto‐medially in the antebrachium superficial to m. extensor carpi radialis longus and medial to m. extensor digitorum communis until its distal end where it converges to insert onto a distinct tubercle on the distal radius proximal to the styloid process (Figure 3).

3.6.2. M. extensor carpi radialis longus

M. extensor carpi radialis longus has a fleshy origin from the cranial surface of the proximal lateral supracondylar ridge of the humerus (Figure 2) and is fused with m. extensor carpi radialis brevis, which lies deep to it. This fusion was substantial in the left limb, but less so on the right limb. The fleshy fibers of this muscle converge into a tendon, pass through an osseofibrous tunnel in the extensor retinaculum, and insert onto the dorsomedial aspect of the base of MC II (Figures 4 and 5).

FIGURE 4.

Left distal radius and ulna illustrating the location of extensor tendons as they pass deep to the extensor retinaculum. Tendon abbreviations: Ab1long; abductor digiti I longus Ed12, extensor digit I et II; Ecrb, extensor carpi radialis brevis; Ecrl, extensor carpi radialis longus; Ecu, extensor carpi ulnaris; Edc, extensor digitorum communis; Edl, extensor digitorum lateralis. [Planned for column width]

FIGURE 5.

Manus muscle map for Panthera tigris (left side). (a) Dorsal view; (b) palmar view. Light gray = origins, dark gray = insertions, black and white striped areas = ligament attachments. Numbers (1–4) correspond the digit served by the muscle; l = lateral head, m = medial head. Muscular abbreviations: Ab1long, abductor digiti I longus; Ab5, abductor digiti V; Abo1, abductor et opponens digiti I; Ad, adductores digitorum; Ecrb, extensor carpi radialis brevis; Ecrl, extensor carpi radialis longus; Ecu, extensor carpi ulnaris; Ee, extensor expansions; Fbp, flexores breves profundi; Fcr, flexor carpi radialis; Fcu, flexor carpi ulnaris; Fdbm, flexor digitorum brevis manus; Fdp, flexor digitorum profundus; Fds, flexor digitorum superficialis; L, lumbrical; Pl, palmaris longus; Pmc, pisometacarpal ligament; X = location of attachment for the tendon joining the deep surface of the tendinous tunnel for m. flexor digitorum profundus of digit II to the palmar surface of the base of the proximal phalanx of digit I. [Planned for page width]

3.6.3. M. extensor carpi radialis brevis

M. extensor carpi radialis brevis originates from a distinct cranially projecting crest on the lateral supracondylar ridge of the humerus distal to m. extensor carpi radialis longus (Figure 2). In the right limb, this muscle was partially fused at its origin with m. extensor carpi radialis longus, while in the left limb, it was also partially fused with m. extensor digitorum communis. The muscle bellies of mm. extensor carpi radialis longus and brevis share fibers for much of their origin and diverge in the mid antebrachium where m. extensor carpi radialis brevis forms a tendon and passes through the extensor retinaculum with m. extensor carpi radialis longus to insert onto the dorsomedial aspect of the base of MC III (Figures 4 and 5).

3.6.4. M. extensor digitorum communis

This muscle originates from the lateral supracondylar ridge of the humerus distal to the origin of m. extensor carpi radialis brevis (Figure 2). The body of the muscle gives rise to four separate tendons for digits II–V, but only the tendon to digit II has its own distinct muscle belly. The tendons pass together through a fibrous tunnel deep to the extensor retinaculum lateral to mm. extensor carpi radialis longus and brevis (Figure 4) to contribute to the dorsomedial extensor expansions of digits II–V.

3.6.5. M. extensor digitorum lateralis

M. extensor digitorum lateralis originates from the lateral epicondyle of the humerus just proximal to the radial collateral ligament (Figure 2) and gives rise to three tendons to digits III–V. The tendons to digits III and IV are tightly connected as they travel deep to the extensor retinaculum and separate in the manus where they join the dorsolateral aspects of the extensor expansions to digits III and IV. Distal to this bifurcation, the tendon to digit III sends a small tendon to join the lateral tendon of m. extensor digiti I et II. The tendon to digit V separates from the main tendon proximal to the wrist and travels through its own osseofibrous tunnel to contribute to the dorsolateral extensor expansion of digit V (Figure 4).

3.6.6. M. extensor carpi ulnaris

A shallow depression with a distinct, raised rim marks the origin of m. extensor carpi ulnaris from the inferior aspect of the lateral condyle of the humerus (Figure 2). The muscle gives rise to a flat, wide tendon just proximal to the wrist joint and passes lateral to the ulnar styloid process deep to, and partially blending with, the extensor retinaculum (Figure 4). It inserts onto the lateral base of MC V and blends with the pisometacarpal ligament. The tendon of insertion for m. extensor carpi ulnaris also provides partial origin for m. abductor digiti V (Figure 5b).

3.6.7. M. supinator

M. supinator lacks a bony origin, originating from the joint capsule and the anular and lateral collateral ligaments. It has a fleshy insertion on the proximal half of the cranial surface of the radial shaft (Figure 3) and to the underlying mm. pronator teres and abductor digiti I longus. Some fibers from this muscle cross the interosseous space proximal to the interosseous membrane to attach to the deep surface of m. abductor digiti I longus.

3.6.8. M. abductor digiti I longus

This muscle has an extensive origin from the entire length of the lateral ulnar shaft extending from just distal to the insertion of m. anconeus (Figure 3). It also originates from the lateral surface of the radial shaft distal to the radial tuberosity and the interosseous membrane. The origin also extends onto the cranial aspect of the distal radius, contacting m. pronator teres at its insertion. Its tendon travels deep to the extensor retinaculum through its own osseofibrous tunnel medial to mm. extensor carpi radialis longus and brevis (Figure 4) and inserts onto a tubercle on the base of MC I and a radial sesamoid (Figure 5b).

3.6.9. M. extensor digiti I et II

This muscle originates from a ridge on the lateral aspect of the caudal surface of the ulnar shaft, caudal to the origin of m. abductor digiti I longus and lateral to the origin of m. flexor digitorum profundus caput ulnare (Figure 2). It gives rise to one tendon that travels deep to the extensor retinaculum lateral to m. extensor digitorum communis and medial to m. extensor digitorum lateralis (Figure 4). Distal to the extensor retinaculum, this tendon passes deep to the four tendons of m. extensor digitorum communis and bifurcates into two tendons to digits I and II. The tendon to digit II joins the extensor expansion for this digit lateral to the tendon of m. extensor digitorum communis and sends a small tendinous slip to the tendon of m. extensor digitorum lateralis to digit III.

3.6.10. Dorsal digital expansions (=extensor expansions)

The long extensor muscles of digits II–V all contribute to the formation of dorsal digital expansions, broad sheet‐like tendons that spread across the dorsal aspect of the digits. The extensor expansion of digit I is not well developed, and is formed mainly by the tendon of m. extensor digiti I et II to digit I. This tendon inserts onto the proximal aspect of the dorsal surface of the distal phalanx of digit I (Figure 5a). The extensor expansions to digits II–V are more elaborate with multiple muscles of the caudal antebrachium and intrinsic manus contributing to them (specific contributions are noted in those particular sections of this paper). The main insertion of the extensor expansions of digits II–V is a tubercle on the proximal aspect of the dorsal surface of the intermediate phalanges (Figure 5a). A smaller slip leaves the extensor expansion near the main insertion and inserts onto the proximal surface of the terminal phalanx just dorsal to the articular surface for the head of the intermediate phalanx.

3.7. Cranial compartment of the antebrachium

This compartment has a larger muscle mass than the caudal antebrachium, comprising 15.0% of the forelimb. M. flexor digitorum profundus contributes 42.7% of the muscle mass of the cranial antebrachium, with the second largest muscle being m. flexor carpi ulnaris, making up 17.97% of muscle mass, and m. palmaris longus being roughly equivalent at 16.63%. M. pronator teres contributes 10.38% to cranial compartment muscle mass and mm. flexor carpi radialis and pronator quadratus each makes up 5.4% and 4.88%, respectively. Mm. flexor digitorum superficialis (1.08%) and flexor digitorum brevis manus (0.96%) are the smallest muscles of the cranial antebrachium.

3.7.1. M. pronator teres

M. pronator teres originates from a distinct facet on the proximal aspect of the medial epicondyle of the humerus and ulnar collateral ligament (Figure 2). It inserts via fleshy fibers onto a rough tubercle on the cranial aspect of the middle third of the radial shaft and to the fascia overlying m. pronator quadratus (Figure 3). Fibers from m. supinator insert onto m. pronator teres at the site of its attachment onto the radius. On the right side, m. pronator teres took partial origin from m. flexor carpi radialis.

3.7.2. M. flexor carpi radialis

This muscle originates from the medial epicondyle of the humerus just distal to m. pronator teres and from the ulnar collateral ligament (Figure 2). The muscle belly remains distinct throughout its length and gives rise to a tendon just before it passes into the wrist deep to the flexor retinaculum and through a groove on the palmar surface of the scapholunar, where it bifurcates to insert onto the palmar bases of MC II and III (Figure 5b).

3.7.3. M. palmaris longus

The palmaris longus originates from a smooth, round facet on the caudal aspect of the medial epicondyle of the humerus (Figure 2). The superficial surface of the muscle belly is covered by an aponeurosis for much of its length which becomes progressively thicker distally to form a thick tendon that serves as the origin for m. flexor digitorum brevis manus just proximal to the wrist. This tendon passes through its own retinaculum superficial to the transverse carpal ligament before dividing into three large tendons to digits II–IV and two much smaller tendons to digits I and V. The tendon to digit I partially blends with the lateral part of the anular ligament at the metacarpophalangeal joint and inserts onto the ulnar sesamoid of digit I (Figure 5b). The tendons to digits II–V pass deep to the anular ligament to ultimately insert onto a distinct fossa on the palmar base of the intermediate phalanx of their corresponding digit (Figure 5b). In digits II–V, the tendons also contribute to the tunnels for m. flexor digitorum profundus together with m. flexor digitorum superficialis (digits II–IV) and m. flexor digitorum brevis manus (digits IV–V). At the level of the metacarpophalangeal joints of digits II–V, transverse fibers connect the tendons of m. palmaris longus to each other. These tendons are also connected to the fatty palmar paw pad by small reticular fibers and by a slender but distinct ligament. An additional tendon joins the deep surface of the tendinous tunnel for m. flexor digitorum profundus of digit II to the palmar surface of the base of the proximal phalanx of digit I (Figure 5b).

3.7.4. M. flexor carpi ulnaris

M. flexor carpi ulnaris consists of two heads, caput humerale and caput ulnare. The caput humerale originates from a smooth area on the caudal aspect of the medial epicondyle lateral to the origin of m. palmaris longus (Figure 2). Proximally, m. flexor carpi ulnaris caput humerale shares fibers with m. flexor digitorum profundus caput humerale laterale. M. flexor carpi ulnaris caput ulnare originates from the caudomedial aspect of the olecranon process. The two heads of m. flexor carpi ulnaris are separate for most of their length, until just proximal to the wrist where they fuse and insert onto the tuberosity of the pisiform via a single tendon (Figure 5b).

3.7.5. M. flexor digitorum superficialis

M. flexor digitorum superficialis is a small muscle in the distal third of the antebrachium that originates from the superficial surface of m. flexor digitorum profundus just proximal to the wrist (Figure 6d). This muscle consists of three heads, serving digits II–IV, each of which gives off a single long tendon. The head serving digit II originates medial to the other two; the head serving digit IV originates deep to the head serving digit III and is partially fused with it at its origin. The three tendons continue into the manus in between the tendons of mm. flexor digitorum profundus and palmaris longus. M. flexor digitorum superficialis tendons to digits II and III remain separate from the tendons of m. palmaris longus until the level of the metacarpophalangeal joint, where they each bifurcate and fuse with the tendon of m. palmaris longus to allow passage of m. flexor digitorum profundus tendon. The tendon to digit IV fuses first to the tendon of m. flexor digitorum brevis manus before the conjoined tendon fuses with that of m. palmaris longus at the level of the metacarpophalangeal joint and bifurcates as in digits II and III. The bifurcated tendons of digits II–IV pass on either side of the tendon of the flexor digitorum profundus to insert onto the proximal base of the intermediate phalanx with the tendons of m. palmaris longus (Figure 5b).

FIGURE 6.

Palmar view of the right (b) and left (a,c,d) manus showing the relationship of mm. palmaris longus, flexor digitorum brevis manus, and flexor digitorum superficialis. FBP, flexor brevis profunda; FDBM, flexor digitorum brevis manus; FDP, flexor digitorum profundus; FDS, flexor digitorum superficialis. [Planned for page width]

3.7.6. M. flexor digitorum profundus

This muscle consists of five heads: capita humerale mediale, humerale laterale, humerale profundus, ulnare, and humerale radiale et ulnare. The caput humerale mediale originates on the medial epicondyle just distal to m. flexor carpi radialis. The caput humerale laterale originates from the distal surface of the medial epicondyle distal and lateral to the caput humerale mediale. The caput humerale profundus originates from a fossa on the distal surface of the medial epicondyle of the humerus to the origins of mm. palmaris longus and flexor carpi ulnaris caput humerale and lateral to the origin of m. flexor digitorum profundus caput laterale (Figure 2). Although their specific origins can be traced to the bone, the tendons of the capita humerale laterale, mediale, and profundus are fused proximally and separate in the mid antebrachium. M. flexor digitorum profundus caput ulnare originates from the medial surface of the olecranon and ulnar shaft. The caput radiale et ulnare originates from the ulnar shaft distal to the insertion of brachialis and cranial to the caput ulnare, the interosseous membrane, and the proximal third of the medial radial shaft (Figure 3).

All bellies fuse at the wrist into a single, flat tendon with decussating fibers, which passes deep to the flexor retinaculum with m. flexor digitorum superficialis, deep to the tendon of m. palmaris longus. In the manus, the common tendon divides into five tendons, one to each digit, each of which inserts onto the flexor tubercle of the distal phalanx (Figure 5b).

3.7.7. M. flexor digitorum brevis manus

M. flexor digitorum brevis manus consists of three distinct heads of origin, two superficial and one deep (Figure 6). The lateral superficial head originates from the superficial surface of the palmaris longus muscle in the distal antebrachium, while the medial superficial head originates from the deep surface of the retinaculum overlying the tendon of m. palmaris longus and from the superficial surface of the retinaculum for m. flexor digitorum profundus. The deep head of m. flexor digitorum brevis manus originates from the deep and lateral surface of m. palmaris longus tendon, and partially fuses with the origin of the lateral superficial belly (Figure 6c). The two superficial heads fuse distally to give rise to a single tendon to digit V, and the deep head provides a tendon to digit IV. In the manus, the tendon to digit IV fuses with that from m. flexor digitorum superficialis before fusing with the tendon from m. palmaris longus. The tendon to digit V fuses with the tendon of m. palmaris longus at the metacarpophalangeal joint. The course of the tendons of mm. palmaris longus and flexor digitorum superficialis is described in detail above.

3.7.8. M. pronator quadratus

M. pronator quadratus has an extensive origin from the distal half of the medial shaft of the ulna and inserts on the distal half of the medial shaft of the radius (Figure 3). Both the origin and insertion of this muscle are marked by distinct bony ridges. The superficial surface of this muscle is covered by the deep antebrachial ligament, which attaches to the radius just proximal to the superior border of m. pronator quadratus and blends with the deep fascia overlying the carpal bones in the carpal tunnel.

3.8. Manus

The muscles of the manus comprise just 1.99% of the mass of the forelimb. Within the manus, m. flexores breves profundi make up approximately 70.28% of total muscle mass, followed by m. abductor digiti V (11.08%), mm. adductores digitorum (9.55%), and mm. lumbricales (8.9%). The smallest muscle in the manus is m. abductor et opponens digiti I, which contributes only 0.19% of the mass of the manus.

3.8.1. M. palmaris brevis

M. palmaris brevis is a small muscle that originates from the flexor retinaculum and deep fascia overlying the lateral palm. It consists of disorganized fibers that mix with fibrous and fatty tissue within the carpal paw pad.

3.8.2. M. abductor et opponens digiti I

This is a small muscle, about 2 cm long, originating from the medial flexor retinaculum overlying m. palmaris longus tendon (Figure 6) and inserting onto the medial sesamoid of the metacarpophalangeal joint of digit I (Figure 5b).

3.8.3. M. abductor digiti V

M. abductor digiti V originates via fleshy fibers from the lateral surface of the tubercle of the pisiform (Figure 5b), the tendons of m. flexor carpi ulnaris and m. extensor carpi ulnaris, the lateral surface of the pisometacarpal ligament to MC V, and the palmar surface of the pisometacarpal ligament to MC IV. On the right side, an additional slip of muscle originated from the flexor retinaculum overlying the tendon of m. flexor digitorum profundus medial to the pisometacarpal ligament to MC IV. This muscle forms a large, fleshy belly that rapidly tapers to form a long tendon that travels along the lateral aspect of the manus to insert on the lateral surface of the base of the proximal phalanx and lateral sesamoid of digit V (Figure 5b). Some fibers also join the lateral extensor expansion.

3.8.4. Mm. lumbricales

Four muscle bellies are present, serving digits II–V. The belly to digit II originates from the lateral base of m. flexor digitorum profundus tendon to digit I and the medial and palmar aspects of the tendon to digit II. Muscle bellies to digits III–V originate from the palmar surface of the common tendon of m. flexor digitorum profundus; m. lumbricals IV and V are also partially fused at their origin. Each muscle belly inserts on the medial base of the proximal phalanx of their respective digits (Figure 5b). They did not contribute to the extensor expansions of the digits.

3.8.5. Mm. adductores digitorum

This muscle group consists of three distinct muscle bellies serving digits I, II, and V. M. adductor digiti I originates from ligaments overlying the insertion of the flexor carpi radialis and inserts onto the lateral base of the proximal phalanx of digit I (Figure 5b).

M. adductor digiti II originates from ligaments overlying the central carpus in common with the medial head of m. adductor digiti V and lateral to the origin of m. adductor digiti I. It inserts onto the lateral base of the proximal phalanx of digit II (Figure 5b).

M. adductor digiti V has distinct medial and lateral heads. The medial head originates with m. adductor digiti II from ligaments overlying the central carpus and the lateral head originates from the medial head of m. flexor brevis profunda to digit V. Muscle fibers from the two heads run distally and insert separately. The lateral head inserts onto a roughened area the distal aspect of the medial shaft of MC V and the medial head inserts onto the medial base of the proximal phalanx of digit V (Figure 5b). In the right limb a long ligament connects the hook of the hamate to the region of insertion of the adductor digiti V lateral head; this ligament is absent in the left limb.

3.8.6. Mm. flexores breves profundi

Each of mm. flexores breves profundi consists of a medial and lateral head except for the one serving digit I, which has only one head.

M. flexor brevis profunda digiti I originates from ligaments overlying the medial carpus and inserts onto the medial aspect of the shaft of MC I and onto the medial sesamoid of the metacarpophalangeal joint of digit I (Figure 5b).

The medial and lateral heads of m. flexor brevis profunda digiti II share a common origin from ligaments overlying the distal carpus and proximal metacarpus and from the tendon of m. flexor carpi radialis. The medial head also has a small origin from the base of MC I and the lateral head has a secondary origin from the proximal shaft of MC II. Fibers from the medial and lateral heads combine and then split again to travel on either side of the metacarpophalangeal joint. Each inserts onto its respective sesamoid bone at the metacarpophalangeal joint (Figure 5b) and contributes to the extensor expansion of digit II. The lateral head also inserts onto the lateral base of the proximal phalanx of digit II.

The medial head of m. flexor brevis profunda digiti III originates from ligaments overlying the central carpal bones and from the proximal bases of MC II and III near the insertion for m. flexor carpi radialis (Figure 5b). The lateral head originates from the bases of MC III and IV in addition to its ligamentous origin. The medial head inserts onto the medial sesamoid and extensor expansion of digit III and the lateral head does the same on the lateral side.

The medial head of m. flexor brevis profunda digiti IV originates from the ligaments overlying the lateral carpus. The lateral head has secondary origins from the medial shaft of MC V and the lateral shaft of MC IV and to the base of MC V medial to the attachment of the os carpi accessorium ligament (Figure 5b). The two heads divide and insert onto their respective sesamoids and to the extensor expansion on either side of digit IV. The lateral head has an additional insertion onto the lateral base of the proximal phalanx.

M. flexor brevis profunda digiti V originates from the base of MC V distal to the attachment of the os carpi accessorium ligament and ligaments of the carpometacarpal joint (Figure 5b). From the single origin, the muscle divides into medial and lateral heads that insert onto the corresponding sesamoid of digit V and contribute to the extensor expansion.

4. DISCUSSION AND COMPARISONS

4.1. Quantitative analyses

The lack of discrimination between locomotor categories in all muscular groups is consistent with other studies that find that felids are extremely conservative in morphology (Cuff et al., 2016; Meachen‐Samuels & Van Valkenburgh, 2009; Smith et al., 2021) and emphasizes the dual role that the forelimb plays in both prey capture/manipulation and locomotion in this group (Cuff et al., 2016; Gonyea, 1976; Gonyea & Ashworth, 1975; Meachen‐Samuels & Van Valkenburgh, 2009). The negative correlation of regional shoulder muscle mass and body mass suggests that smaller cats have relatively larger shoulder muscles than larger cats. When muscles acting on the shoulder are considered in functional groups, neither shoulder extensors nor shoulder flexors shows a significant relationship to body mass. However, both shoulder flexors and extensors are significantly larger in felines than pantherines, which is not surprising considering felines are, on average, smaller than pantherines. Functionally, this suggests that larger cats have relatively weaker shoulder muscles than smaller cats; however, it is difficult to determine why this might be the case. This is somewhat contradictory to Cuff et al. (2016), who found that some extrinsic shoulder muscles and the infraspinatus scaled with positive allometry in felids; however, that study also found that overall, individual muscles of the forelimb scaled with geometric similarity in felids with or without phylogenetic correction, also suggesting that larger cats are relatively weaker than smaller cats.

The caudal brachium showed the opposite pattern, being positively correlated with body mass and with felines having significantly less massive muscles of this compartment than pantherines. As most of these muscles are elbow extensors, more massive muscles in this region may be related to the need for increased resistance of gravity in a larger animal. A larger caudal brachium may also correlate with the use of forelimbs in larger cats to subdue large prey. Histophysiological studies on both cats and dogs have found that deep muscles within the caudal brachium (e.g., mm. anconeus, triceps brachii caput mediale) are primarily composed of slow‐twitch muscle fibers and primarily function in joint stability and resistance to gravity, whereas more superficial muscles of this group (e.g., mm. triceps brachii caput longum and laterale) comprise largely fast‐twitch fibers and perform faster or more powerful contractions (Armstrong et al., 1982; Collatos et al., 1977). Thus, either one or both of these factors could be occurring depending on which individual muscles are driving these differences. Our analysis shows a significant difference in the mass of functional elbow extensors, with large prey specialists having larger elbow extensors and mixed or small prey specialists having smaller elbow extensors. This makes functional sense, as elbow extension is one of the main modes of stabilizing a large prey animal, which can also be detected in the size of the olecranon process (Meachen‐Samuels & Van Valkenburgh, 2009). The lack of significant differences in prey specialists for the caudal brachium group as a whole is likely due to the smaller sample size, since there were fewer taxa for which individual muscle masses were available.

Muscles of the cranial antebrachium demonstrated a positive correlation with body mass in our felid sample, indicating that larger species had relatively larger muscles. This may be related again to prey capture, as larger felids tend to take larger prey and need to hang on to it. However, this is not fully supported by our analysis of functional muscle groups, which found no significant differences in carpal or digital flexors or extensors, nor any significant correlation with body mass. Of the functional groups of the forearm, only supinators and pronators showed significant associations; both of these groups were positively correlated with body mass and were significantly larger in pantherines than felines. Cuff et al. (2016) found that m. brachioradialis (a supinator) scaled with positive allometry, which may be driving our result. In fact, a post hoc test for differences in the relative mass of m. brachioradialis revealed a significantly larger muscle in pantherines versus felines, whereas no significant difference was found for m. supinator. M. brachioradialis makes up less than 0.5% of total forelimb mass in all felines with the exception of the ocelot and the cheetah. Again, the larger size of these muscles may be related to differences in prey preferences in smaller versus larger cats and the fact that no significant differences between prey categories were found in our analyses may simply be due to the small sample and uneven distribution of taxa belonging to the three different prey preference categories (Tables 4 and 5).

4.2. Muscle attachments and mass

In general, the origins, insertions, and proportional muscle mass is remarkably similar across felid taxa, suggesting that felids are quite conservative in muscular anatomy across a wide range of body sizes, locomotor behaviors, and prey preferences. In this section, we highlight some of the differences among felids. We present functional interpretations when they are well‐supported by biomechanical principles; however, many of the differences we observe do not have an obvious functional explanation and may simply be due to individual variation. Unless otherwise noted, muscle mass is given as a percentage of total upper limb mass as calculated from the muscles used in the regional groups analysis rather than total mass of the upper limb including all muscles as is given in Table 1 to facilitate comparison with as many taxa as possible.

4.2.1. Extrinsic and intrinsic shoulder muscles

The insertion of m. omotransversarius is expanded in the tiger, extending from the tip of the suprahamate process of the scapula onto the proximal scapular spine, whereas in the snow leopard, ocelot, and domestic cat the insertion is restricted to the extreme tip of the suprahamate process (Julik et al., 2012; Reighard & Jennings, 1901; Smith et al., 2021). The bony origin of m. deltoideus pars scapularis is restricted to the caudal tip of the suprahamate process, resembling the condition in the ocelot (Julik et al., 2012) and differing from that seen in the snow leopard and domestic cat (Reighard & Jennings, 1901; Smith et al., 2021), in which this muscle takes extensive origin from the mid‐portion of the scapular spine. This muscle also originates from the fascia overlying m. infraspinatus in P. uncia and L. pardalis (Julik et al., 2012; Smith et al., 2021) and it is unclear what, if any, functional significance this difference in bony origin would have.

On both right and left limbs, we observed m. infraspinatus originating from the tendon of origin of m. triceps brachii caput longum. This particular configuration has not been noted in other felids, however, in his description of the muscular anatomy of the forelimb of the tiger, Scharlau (1925) described m. triceps brachii caput longum as originating in part from the fascia of m. infraspinatus, which may be describing the same feature from a different perspective. Similarly, Davis (1949) describes the posterior portion of m. triceps brachii caput longum as originating from the fascia covering mm. subscapularis minor, infraspinatus, and teres major. The underlying cause of this difference is difficult to discern. We considered that m. infraspinatus may have an expanded origin due to it being proportionally large; however, it is only slightly above the average of pantherines in our sample at 8.60% of forelimb mass (pantherine average = 8.37%) and well below the feline average (9.70%) and overall sample average (9.17%). The bony origin of m. teres minor was restricted to a small region on the caudal border of the scapular neck in the tiger, resembling the domestic cat (Reighard & Jennings, 1901), whereas it extends farther dorsally in the snow leopard and ocelot (Julik et al., 2012; Smith et al., 2021). We also found this muscle to have a secondary origin from the proximal tendon of m. triceps brachii caput longum next to the origin of m. infraspinatus.

The insertion of m. teres major in the tiger resembles that of the lion, being more distally positioned on the medial humeral shaft and terminating at or near the midshaft of the humerus (Figure 2), rather than terminating well proximal to midshaft as in the snow leopard, cheetah, Eurasian lynx, ocelot, and domestic cat (Hudson et al., 2011; Julik et al., 2012; Reighard & Jennings, 1901; Smith et al., 2021; Viranta et al., 2016). The distal position of this insertion relative to the glenohumeral joint increases the lever‐arm for this muscle, making it more powerful in shoulder extension compared with taxa with more proximal insertions, in which contractions of this muscle would produce faster, but less powerful extension (Dunn, 2018; Hildebrand et al., 1985). The latter condition is seen in the cheetah, which relies on rapid terrestrial pursuit for prey capture (Sunquist & Sunquist, 2002). In contrast, the more powerful shoulder extension in the lion and tiger may provide them more effective control of large prey moving in haphazard directions. The shoulder extensors, in conjunction with digital and wrist flexors, allow these big cats to keep the prey close to their body for better control. This is especially true when the big cat holds on to the prey on the ground for the killing bite.

The distribution of muscle mass among intrinsic shoulder muscles shows only minor variation among felids. P. tigris has the largest mm. supraspinatus, and subscapularis of the pantherines. M. supraspinatus makes up approximately 12% of total forelimb mass, although the differences from the other taxa are small (10.5% in P. leo, 10.8% in P. onca, and 11.7% in P. uncia). The cheetah has the largest supraspinatus of all felids, with this muscle making up 16% of the mass of the forelimb. It is possible that the relatively large supraspinatus in the tiger and cheetah is related to their predominately terrestrial mode of locomotion as this muscle contributes to flexion of the shoulder during forward movement. This is supported by a relatively large m. supraspinatus in the terrestrial domestic cat (11.9%); however, this muscle is also relatively large in the caracal (13.2%), which is considered scansorial. M. subscapularis represents 9.35% of the mass of the forelimb in P. tigris, with the next‐largest pantherine being 9.29% in P. onca. P. uncia and P. leo both have a smaller m. subscapularis at 8.54% and 7.34%, respectively. Among felids, F. nigripes (12.4%), L. pardalis (11.7%), and F. catus (10.3%) have the largest m. subscapularis and L. lynx has the smallest (8.1%). It is difficult to find a common trait linking the taxa with a large m. subscapularis together as the list includes large and small prey specialists as well as terrestrial and scansorial taxa from felines and pantherines. Hudson et al. (2011) also report relatively large mm. supraspinatus, infraspinatus, and subscapularis in the cheetah relative to the similarly sized cursorial greyhound and note that shoulder muscles may be larger in the cheetah in part to counter peak reaction forces while maneuvering at high speeds during pursuit of prey.

4.2.2. Caudal compartment of the brachium

The origin of m. triceps brachii caput longum is expanded in the tiger, originating from almost the entire caudal border of the scapula. This is similar to the condition in the lion described by Julik et al. (2012) and Davis (1949); Barone (1967) likewise described this muscle as originating from the distal two‐thirds of the scapula in the lion, and it appears to be similarly extensive in the cheetah (Hudson et al., 2011). In contrast, m. triceps brachii caput longum originates from the distal half of the caudal edge of the scapula in the snow leopard, European lynx, and ocelot (Julik et al., 2012; Smith et al., 2021; Viranta et al., 2016), and from the distal third in the domestic cat (Reighard & Jennings, 1901). It is not particularly surprising that P. tigris would resemble P. leo in this regard as these species are closely related and share mostly terrestrial locomotor behavior and a preference for large prey. Although P. uncia, L. pardalis, and L. lynx differ in body size and habitats, all are broadly classified as scansorial and so may not benefit from greater development of m. triceps brachii caput longum as an elbow and shoulder extensor active in terrestrial pursuit (Julik et al., 2012; Meachen‐Samuels & Van Valkenburgh, 2009; Smith et al., 2021; Sunquist & Sunquist, 2002; Viranta et al., 2016). The similarly extensive origin of this muscle in the highly cursorial A. jubatus is consistent with this muscle being important for rapid terrestrial locomotion (Hudson et al., 2011). Davis (1949) suggested that the enlarged m. triceps brachii caput longum in Ursus americanus functioned mainly to resist forward pull (flexion) of the forelimb in climbing; likewise in these large terrestrial felids, the more caudal expansion of the origin for this muscle results in a larger moment arm of the muscle at the glenohumeral joint and thus more power in extension of the shoulder. M. triceps brachii caput longum is the largest head of m. triceps brachii by mass in all felid taxa for which data are available. It is largest in the cheetah, at 19.8% of the mass of the forelimb and smallest in the ocelot at 13.9%. On average, this muscle is larger in pantherines (17.3%) than felines (16.3%); however, there is substantial overlap between the two groups. Although the origin of m. triceps caput longum is expanded in the tiger, this muscle is small compared with other pantherines at 16.43% of overall forelimb mass.

M. triceps brachii caput laterale originates more distally on the humeral shaft in the tiger, resembling the origin of this muscle in P. uncia, rather than being located just distal to the head as in the ocelot, and domestic cat (Julik et al., 2012; Reighard & Jennings, 1901; Smith et al., 2021).The origin of m. triceps brachii caput mediale is more extensive in the tiger than in P. uncia, L. pardalis, or F. catus, resembling L. lynx in extending distally past midshaft rather than only extending to midshaft as in the other taxa. M. triceps brachii caput mediale accessorium originates from the entire caudal surface of the medial epicondyle and medial supracondylar ridge in the tiger (Figure 2) and snow leopard (Smith et al., 2021), whereas in the other felid its origin is restricted to the medial supracondylar ridge (Julik et al., 2012; Reighard & Jennings, 1901; Smith et al., 2021; Viranta et al., 2016). The larger m. triceps brachii caput longum, and to a lesser extent caput laterale, in P. tigris and P. leo may reflect a greater emphasis on capturing larger prey while using the forelimbs to push the prey down on the ground before and during the killing bite. This is consistent with other studies that found that large‐prey specialists tend to have larger forearm extensor muscles than small‐prey specialists (Smith et al., 2021; Souza Junior et al., 2021).

The tiger has the largest m. triceps brachii caput laterale of the pantherines at 7.13% of the mass of the forelimb, with the lion having the smallest (6.22%). The relatively large size of this muscle in the tiger may reflect an emphasis on terrestrial pursuit or grappling with large prey, as this muscle has been shown to be composed primarily of fast‐twitch, Type II muscle fibers in both cats and dogs (Armstrong et al., 1982; Collatos et al., 1977). The largest m. triceps brachii caput laterale of our sample is found in the cheetah (7.65%). The extensive origins of mm. triceps brachii caput mediale is not reflected in an enlarged muscle mass, with the tiger (2.05%) being below the average for pantherines (2.44%) and felines (2.12%).

M. anconeus in the tiger is relatively simple compared with many other felids. It takes origin from the distal quarter of the caudal humeral shaft, similar to the condition in the cheetah (Hudson et al., 2011) and that described in the snow leopard, although the muscle maps show this muscle originating from the distal half of the humerus (Smith et al., 2021). In the domestic cat, it originates from the distal third of the humeral shaft and from the distal half in the ocelot (Julik et al., 2012; Reighard & Jennings, 1901). M. anconeus consists of one head in the tiger and domestic cat (Reighard & Jennings, 1901), whereas in the lion and snow leopard, it is divisible into two distinct heads (Barone, 1967; Smith et al., 2021). Two to three heads are also present in the ocelot with two to three sesamoids present in its tendon of insertion (Julik et al., 2012). The insertion of m. anconeus in the tiger is proximodistally restricted as compared with the snow leopard and ocelot but is cranio‐caudally expanded, covering much of the lateral surface of the olecranon process rather than being restricted to a thin, elongate ridge. (Julik et al., 2012; Smith et al., 2021). Physiological studies on the muscular composition of m. anconeus in cats and dogs show that this muscle is composed of close to 100% slow‐twitch muscle fibers, suggesting that the primary function of this muscle is to resist gravity and stabilize the elbow joint (Armstrong et al., 1982; Collatos et al., 1977). The two felids in the sample showing the largest m. anconeus are the snow leopard (1.41%) and ocelot (1.39%), while the felid with the smallest is the cheetah (0.31%). Based on the fact that the two largest muscles belong to scansorial cats and the smallest to a terrestrial cat, it is tempting to infer that m. anconeus may play a role in elbow stabilization on uneven substrates. However, this pattern does not hold up across the sample, with the scansorial jaguar having a relatively small muscle (0.89%) and the terrestrial tiger having a larger one (1.07%).

4.2.3. Cranial compartment of the brachium

In the tiger, m. articularis humeri originates from the coracoid process, similar to the lion, ocelot, domestic cat, and cheetah (Barone, 1967; Hudson et al., 2011; Julik et al., 2012; Reighard & Jennings, 1901; Scharlau, 1925). In the snow leopard, this muscle is described as originating from the medial margin of the glenoid fossa, although the muscle map shows it originating from the scapular neck just dorsal to the origin of m. biceps brachii (Smith et al., 2021). Viranta et al. (2016) describe this muscle as originating from the coracoid process with a secondary origin from the supraglenoid tubercle. The insertion of m. biceps brachii onto the radial tuberosity and investing fascia of the antebrachium is similar to the condition in most other felids, in contrast to the snow leopard and Eurasian lynx, where this muscle inserts onto both the radius and ulna (Smith et al., 2021; Viranta et al., 2016). The distribution of muscle mass within the cranial brachium is unremarkable, with the tiger being very similar to other pantherines.

4.2.4. Caudal compartment of the antebrachium

The attachments of m. brachioradialis in the tiger resemble that of the domestic cat, and other generalized carnivores such as the lesser grison and red panda (Ercoli et al., 2015; Fisher et al., 2009; Reighard & Jennings, 1901). The snow leopard and ocelot both show an expanded insertion of this muscle onto the flexor retinaculum and tendon of m. abductor digiti I longus; in addition, the origin of this muscle is fused with m. brachialis in the snow leopard (Julik et al., 2012; Smith et al., 2021). Despite not having an expanded origin or insertion, m. brachioradialis is relatively large in the tiger. M. brachioradialis in the tiger comprises 1.34% of the total forelimb, similar to the lion (1.38%). The only taxon in the sample with a larger m. brachioradialis is the jaguar (1.61%). A large m. brachioradialis in these taxa is consistent with preference for large prey, as this muscle is important in subduing and grappling with prey (Cuff et al., 2016; Gonyea & Ashworth, 1975). However, the snow leopard also prefers larger prey but has a much smaller m. brachioradialis (0.78%). Cuff et al. (2016) noted that m. brachioradialis scaled with positive allometry for both mass and physiological cross‐sectional area (PCSA) before phylogenetic correction and it appears that this muscle is positively correlated with body mass in our sample (p = 0.005, rho = 0.8303) and is significantly larger in pantherines than felines (p = 0.03).

The origin of m. extensor digitorum communis is more extensive in the tiger, originating from the lateral epicondyle and supracondylar ridge, although this muscle is relatively small in the tiger (1.20%) compared with other felids, with only the cheetah having a smaller one (1.08%). In the ocelot and snow leopard, this muscle originates from a small area on the supracondylar ridge alone; however, the size of this muscle differs greatly in these two taxa, with the ocelot having a relatively small one (1.17%) and the snow leopard having the largest in our sample (1.69%). We found that m. extensor digitorum communis divides into four tendons serving digits II–V, as did Scharlau (1925). This pattern is relatively common in felids, being present in the lion (Barone, 1967), ocelot (Julik et al., 2012), and domestic cat (Reighard & Jennings, 1901), as well as in the lesser grison and red panda (Ercoli et al., 2015; Fisher et al., 2009). In P. uncia and L. lynx, however, there is a fifth tendon serving digit I in addition to the ulnar four tendons (Smith et al., 2021; Viranta et al., 2016).

There is some variation in the origin of m. supinator in carnivores. Scharlau (1925) describes m. supinator of the tiger as originating from the lateral epicondyle of the humerus and radial collateral ligament, similar to the configuration in the Eurasian lynx described by Viranta et al. (2016). Barone (1967) also describes a bony origin of this muscle from the lateral epicondyle in the lion. However, we were unable to discern a bony origin for this muscle in either limb of our tiger specimen, finding only an origin from the radial collateral ligament and joint capsule of the elbow. This appears to be the most common configuration in felids, as the snow leopard, jaguar, ocelot, and domestic cat also lack a bony origin for m. supinator (Julik et al., 2012; Reighard & Jennings, 1901; Sánchez et al., 2019; Smith et al., 2021), as do the lesser grison and red panda (Ercoli et al., 2015; Fisher et al., 2009).

Among pantherines, P. leo and P. tigris have the relatively smallest m. supinator, comprising 0.50% of the total muscle mass of the forelimb in the tiger and 0.41% in the lion. P. uncia has the largest m. supinator (0.65%), with the jaguar being a close second at 0.63%. A relatively small supinator muscle may reflect the reduced demand for supination of the forearm in terrestrial locomotion compared with more scansorial behavior. The extremely reduced condition of m. supinator in the cheetah (0.14%) is consistent with this explanation, as is the large m. supinator in the ocelot (0.69%), with the largest in our sample. However, the terrestrial black‐footed cat also has a relatively large m. supinator in the sample (0.64%), again suggesting a more complex explanation for these differences in relative muscle mass.

In the tiger, the origin of m. abductor digiti I longus extends almost the entire length of the lateral surface of the ulna, from caudal to the semilunar notch to just proximal to the styloid process, as well as approximately two‐thirds of the length of the lateral radial shaft (Figure 3). In the ocelot and Eurasian lynx, the origin of this muscle is also quite extensive, but it does not extend as far proximally on the ulnar shaft and is restricted to the middle third of the radial shaft; the origin of this muscle is even less extensive in the snow leopard (Julik et al., 2012; Smith et al., 2021; Viranta et al., 2016). Similarly, the origin of m. extensor digiti I et II is more distally extensive in the tiger, extending almost the entire length of the ulna distal to the semilunar notch, whereas it terminates around midshaft in the ocelot, and is restricted to a small region on the proximal ulna in the snow leopard. The origin is also shifted caudally compared with the snow leopard and ocelot, occupying the caudo‐lateral border of the olecranon shaft rather than the lateral surface of the ulna (Julik et al., 2012; Smith et al., 2021).

M. abductor digiti I longus is relatively larger in the tiger and snow leopard than other pantherines, comprising 1.21% of total forelimb mass in P. tigris and 1.23% in P. uncia and is relatively the smallest in the jaguar (0.34%). The ocelot has the largest m. abductor digiti I longus in the sample (1.28%), while the cheetah has the smallest (0.26%). As in other muscles, the area of bony origin covered by the muscle does not seem to correlate with its relative size; however, a large m. abductor digiti I longus may be an advantage in spreading the paws wide when taking down large prey.

4.2.5. Cranial compartment of the antebrachium

Individual muscle data for the cranial antebrachium are inconsistently available. Many sources group muscles together (e.g., the “flexor digitorum complex” of Cuff et al., 2016), whereas other muscles of the forearm are small and often not reported (e.g., mm. flexor digitorum superficialis, flexor digitorum brevis manus, etc.). Consequently, it is difficult to compare like data across felid taxa for this forelimb compartment, and the only taxa we have comparative data for all of the individual muscles are P. leo and L. pardalis reported by Julik et al. (2012) and our specimen of P. tigris. When possible, we make comparisons with other taxa here as well.

M. pronator teres inserts more proximally on the radius in the tiger, approximately at the middle third of the cranial radial shaft (Figure 3), resembling the condition in the lion (Barone, 1967; Julik et al., 2012). This muscle also inserts at or proximal to the midshaft of the radius in the Eurasian lynx, domestic cat, and cheetah (Hudson et al., 2011; Reighard & Jennings, 1901; Viranta et al., 2016). In the ocelot, snow leopard, and jaguar, however, the insertion of m. pronator teres extends more distally on the radius and passes into the wrist to fuse with the flexor retinaculum (Julik et al., 2012; Sánchez et al., 2019; Smith et al., 2021). In the ocelot and snow leopard, m. pronator teres has a secondary bony insertion onto the radial sesamoid (Julik et al., 2012; Smith et al., 2021). These differences in insertion likely correspond to differences in locomotion, with more proximal insertions in the terrestrial felids and more distal insertions in scansorial and arboreal taxa (Julik et al., 2012; Sánchez et al., 2019; Smith et al., 2021). In contrast, the origin and insertion of m. pronator quadratus are expanded in the tiger relative to the ocelot and snow leopard, covering the distal half of the medial surface of both the radial and ulnar shafts as in the lion, rather than the distal third (Julik et al., 2012; Smith et al., 2021). In the Eurasian lynx, the origin of m. pronator quadratus from the ulna appears to be reduced and the insertion expanded compared with that of the tiger (Viranta et al., 2016).

The cheetah has the smallest m. pronator teres of all felids in our sample at 0.82% of the mass of the forelimb, while the ocelot has the largest (2.0%). In general, it appears that scansorial taxa tend to have a larger m. pronator teres than terrestrial taxa; however, this is not always the case. Pantherines follow the expected pattern with P. onca (1.88%) and P. uncia (1.66%) having the largest m. pronator teres whereas P. tigris (1.63%) and P. leo (1.58%) have slightly smaller muscles. Interestingly, the tiger (1.06%) and lion (0.78%) both have the largest m. pronator quadratus in the sample, with the snow leopard having the smallest among pantherines (0.33%) and the cheetah the smallest overall (0.26%).

M. flexor carpi ulnaris caput ulnare exhibits a reduced and caudally shifted origin in the tiger and snow leopard compared with the ocelot and Eurasian lynx. In the tiger and snow leopard, the origin of this muscle is restricted to a crest on the caudo‐medial aspect of the olecranon process (Figure 3), whereas it originates from much of the medial surface of the olecranon process in the ocelot and lynx (Julik et al., 2012; Smith et al., 2021; Viranta et al., 2016). P. tigris is similar to P. uncia and the mustelid Galictis cuja in having the insertion of m. flexor carpi ulnaris carried to MC IV and V via the pisometacarpal ligament (Ercoli et al., 2015; Scharlau, 1925; Smith et al., 2021); however, P. uncia is the only felid to report a separate insertion of m. flexor carpi ulnaris onto the base of MC V. Perhaps this difference in P. uncia reflects the need for flexible feet when maneuvering rocky terrain (Sunquist & Sunquist, 2002).

There is variation among felids in the insertion of m. flexor carpi radialis. Scharlau (1925) reports that m. flexor carpi radialis has a single tendon of insertion onto the base of MC II, similar to the condition reported in the snow leopard, lion, and Eurasian lynx (Barone, 1967; Smith et al., 2021; Viranta et al., 2016). However, in both limbs of our tiger, the distal tendon bifurcated to insert onto both MC II and III, as has been described in the ocelot, domestic cat, and cheetah (Hudson et al., 2011; Julik et al., 2012; Reighard & Jennings, 1901), although Smith et al. (2021) reported this condition in one limb of their snow leopard sample. M. flexor carpi radialis also inserts onto both MC II and III in the red panda and lesser grison (Ercoli et al., 2015; Fisher et al., 2009).

Among pantherines, m. flexor carpi radialis is largest in P. onca (1.22%), one of the most arboreal taxa in the sample. M. flexor carpi radialis is similar in size in P. uncia (0.89%) and P. leo (0.87%), whereas that of P. tigris (0.71%) is the smallest. Among felines, the terrestrial cheetah has the smallest m. flexor carpi radialis (0.53%) and the scansorial domestic cat has the largest (1.49%); however, the other taxa do not sort clearly into locomotor groups. For m. flexor carpi ulnaris, pantherines have larger ones than felines with the jaguar again having the largest in the sample (5.33%) and the cheetah the smallest (1.37%). The jaguar appears to be an outlier, with an extremely enlarged muscle relative to the rest of the sample, with all other taxa falling between 1.37% and 3.48% of forelimb muscle mass. P. uncia has the smallest m. flexor carpi ulnaris of the pantherines (2.48%). These enlarged muscles may reflect a greater emphasis on carpal flexion in the more arboreal P. onca, although it should be noted that other scansorial taxa do not consistently show enlargement of these two muscles relative to terrestrial taxa.

M. palmaris longus shows variation in the number of tendons and digits it serves. In the domestic cat, the number of tendons ranges from three to five (Reighard & Jennings, 1901), whereas differing numbers have also been reported from the ocelot: Julik et al. (2012) reported four tendons; Sánchez et al. (2019) reported five. This muscle is described as having four tendons for digits II–V in the snow leopard, South American wildcat, and Lion (Barone, 1967; Sánchez et al., 2019; Smith et al., 2021). We found five tendons of m. palmaris longus in both limbs of our tiger specimen, which is consistent with Scharlau's (1925) description. The Eurasian lynx and jaguar are also reported to have five tendons from m. palmaris longus (Sánchez et al., 2019; Viranta et al., 2016).

M. flexor digitorum profundus makes up the largest proportion of the cranial antebrachium in all felid species for which data are available (P. tigris, P. leo, A. jubatus, L. lynx, and L. pardalis); P. tigris has the largest (7.48%) followed by L. lynx (6.62%) and P. leo (5.10%), and A. jubatus (4.46%) and L. pardalis (4.43%) the smallest. The relatively small size of this muscle in A. jubatus makes sense in light of their prey‐killing strategy that does not involve gripping with the digits, demonstrated by the reduced ability in the cheetah to retract and protract its claws. It is possible that the larger size of m. flexor digitorum profundus in the tiger, lynx, and lion reflects the preference for large prey in these taxa and the necessity to hang on while wrestling them to the ground, in contrast to L. pardalis, which takes mixed prey.

Two small muscles in the superficial forearm, mm. flexor digitorum superficialis and flexor digitorum brevis manus are also highly variable in the number of muscle bellies and the number of tendons they form. Although it is difficult to discern what, if any, functional implications this has, we consider it worthwhile to document this variation. Sources report two to three bellies of m. flexor digitorum superficialis in felids, with two bellies being reported in the tiger by Scharlau (1925) as well as in the snow leopard, lion, and domestic cat (Barone, 1967; Reighard & Jennings, 1901; Smith et al., 2021). We found three bellies of m. flexor digitorum superficialis (Figure 6), resembling the condition described in the ocelot (Julik et al., 2012). Likewise, either two or three tendons are present in felids; we found three tendons in our dissection, serving digits II–IV, which is consistent with previous reports from the tiger and lion (as reported in Scharlau, 1925), and ocelot (Julik et al., 2012). A variable number of tendons has been described in both the domestic cat and snow leopard, with either two or three being present (Reighard & Jennings, 1901; Smith et al., 2021).

We found three distinct muscle bellies of m. flexor digitorum brevis manus, two superficial and one deep, similar to the condition described by Julik et al. (2012) in the ocelot. Reighard & Jennings (1901, p. 181) describe this muscle (which they call the: ulnar part of m. flexor sublimus digitorum) as giving rise to two or three tendons each “from distinct slips of muscle” and note an “accessory slip” contributing to the tendon to digit IV or V that originates from the radial side of the flexor retinaculum (probably corresponding to our medial superficial belly). It is not clear if each of these slips should be considered a separate belly, but if so, this muscle would have three in an individual with two tendons (one for each tendon, plus the accessory slip) or four in an individual with three separate tendons. On the other hand, one could consider the main muscle mass giving off the tendons as a single belly and the accessory slip as another, for a total of two. Two distinct bellies have previously been reported in the tiger (Scharlau, 1925), lion (Barone, 1967), and snow leopard (Smith et al., 2021). This muscle most commonly terminates in two tendons serving digits IV and V, as in the tiger, lion, and ocelot (Barone, 1967; Julik et al., 2012; Scharlau, 1925); two to three tendons have been documented in the domestic cat (Reighard & Jennings, 1901), and Smith et al. (2021) recorded either one, two, or four tendons stemming from this muscle in the snow leopard.

4.2.6. Manus

Although there are extensive minor variations in muscle attachments in the manus, the general pattern is overwhelmingly similar in felids. There are, however, a few notable differences. We found a distinct m. abductor et opponens digiti I in both limbs (Figure 6), which is absent in several felid taxa, including the snow leopard, Eurasian lynx, and ocelot (Julik et al., 2012; Smith et al., 2021; Viranta et al., 2016). This muscle is reported as present in the lion (Barone, 1967; Julik et al., 2012) and its presence appears to be variable in the domestic cat, with Julik et al. (2012) reporting it as absent and Reighard & Jennings (1901) describing it as reduced in size.

M. abductor digiti V shows variation in origin across taxa. In the tiger, this muscle originates from the tendons of both mm. flexor carpi ulnaris and extensor carpi ulnaris. However in the snow leopard and ocelot, this muscle originates from only one of these tendons: in the snow leopard it originates from m. flexor carpi ulnaris, and in the ocelot it originates from the extensor carpi ulnaris. This muscle originates from the pisiform in all felids except for the snow leopard (Barone, 1967; Julik et al., 2012; Reighard & Jennings, 1901; Scharlau, 1925; Smith et al., 2021); an origin from the pisiform is also documented in the red panda and lesser grison (Ercoli et al., 2015; Fisher et al., 2009).

We only have comparative data for muscles of the manus in the ocelot and tiger (Julik et al., 2012). However, the tiger has the largest m. abductor digiti V of these taxa, making up 11.08% of the mass of the manus, compared with 9.92% in the tiger and 4.75% in the ocelot. It is possible that the smaller muscle in the ocelot is related to this difference in prey preference, with the ocelot preferring small to medium prey, which may take less manual strength to bring down. Although there are no data indicating the relative size of this muscle in the snow leopard, it is possible that the origin of this muscle from multiple tendons in the tiger, versus just one in the snow leopard and ocelot, corresponds to greater development of this muscle; however, it is unclear functionally why this muscle may be better developed in the tiger than in the snow leopard as both are large prey specialists that use their paws in similar ways.

M. adductor digiti V consists of two insertions in the majority of felids we surveyed, one on the radial base of the proximal phalanx of digit V, and one on the radial surface of the shaft of MC V. This second insertion onto MC V is documented in the tiger (Scharlau, 1925), lion (Barone, 1967), ocelot (Julik et al., 2012), and domestic cat (Reighard & Jennings, 1901). This secondary attachment in the ocelot and domestic cat extends for the entire length of the metacarpal shaft (Julik et al., 2012; Reighard & Jennings, 1901), whereas it is restricted to a rough tubercle just distal to midshaft in the tiger (Figure 5). This secondary insertion is not reported in the snow leopard (Smith et al., 2021). Again, it is possible that the expanded insertion of this muscle in the ocelot corresponds to an enlarged muscle mass, as m. adductor digiti V makes up 9.87% of the mass of the manus in the ocelot, and just 4.54% in the tiger, with the lion being similar at 4.24%.

5. CONCLUSIONS

Felidae are well‐known for being a morphologically conserved group and the forelimb musculature seems to follow that pattern (Cuff et al., 2016; Gonyea, 1976; Julik et al., 2012; Souza Junior et al., 2021). Although our dissection corroborates this general morphological conservatism, our results also demonstrate differences between tigers and other felids that are likely due to body size and prey size preference in large versus medium‐sized felid species, as well as phylogeny. As in other studies (Meachen‐Samuels & Van Valkenburgh, 2009; Smith et al., 2021), our statistical analysis did not show any significant differences in muscle mass between locomotor groups (scansorial/arboreal vs. terrestrial); however, this does not mean that locomotor behavior does not influence muscular anatomy. Interpretation of the functional significance of the muscular data we collected is complicated by a number of factors. First, behavioral categories that are used here and in other studies are necessarily broad and encompass a wide range of specific behaviors that may be driving anatomical variation. While it is true that the lion, tiger, and cheetah are predominantly terrestrial and take predominantly large prey, there are striking differences in the way these taxa pursue prey, and the type of large prey they prefer. These behavioral differences may drive differences in anatomy that would be obscured in a study such as this, in which all of these taxa fall into the same broad categories (Sunquist & Sunquist, 2002).

One other issue that may affect our results involving body mass is that several of our body masses came from species averages rather than individual masses. Considering the pronounced sexual dimorphism in many cat species, sex mass averages are preferable to total species averages, but this distinction is not always available in the literature, and we used the best estimates available to us. Considering that we used available species averages for some taxa, there is likely some error introduced into our calculations of muscle sizes relative to mass. However, this complication could not be avoided, and given that the species compared here span a wide range of masses, these effects should not overshadow the true differences in the muscle groups between taxa.

Another confounding factor is the sheer complexity of the system. Variables such as muscle mass, and location and area of bony origin and insertion undoubtedly influence behavior; however, it is really the interaction of these variables together with a number of other factors, such as fiber‐type composition, soft‐tissue connections, etc., that are responsible for producing behaviors. Thus, differences that we identify may not have an obvious correlation with a behavior, but they may still impact it in some way. This highlights the importance of documenting muscular anatomy and variations in anatomy for the benefit of future research even if no functional significance is immediately evident. Further, anatomical dissection is an extremely time‐consuming process, leading to small samples, which complicates our ability to observe patterns and limits our understanding of inter‐ and intraspecific variation.

Despite these limitations, several features of the muscular anatomy of the tiger are consistent with an emphasis on a predominantly terrestrial locomotor repertoire, including the relatively large mass of m. supraspinatus, a shoulder flexor, and more distal insertion of m. teres major. The expanded origins of m. triceps brachii caput longum and the large muscle mass of m. triceps brachii caput laterale, both elbow extensors may also represent the strong role these muscles play in both terrestrial locomotion and subduing large prey.

Preference for large prey in the tiger may also be reflected in the relatively large mm. brachioradialis, abductor digiti I longus, flexor digitorum profundus, and abductor digiti V. Conversely, the more proximal insertion of m. pronator teres and lack of a secondary bony insertion for this muscle together with a fairly unremarkable “average” muscle mass of the cranial antebrachial muscles suggest a lack of functional specialization for climbing behaviors in the tiger, also reflecting a more terrestrial way of life. Interestingly, tigers resemble P. leo in retaining m. abductor et opponens digiti I (probably the primitive condition in felids) but also resemble the scansorial felids P. uncia and L. pardalis in having only one head of m. flexor brevis profunda for digit I (Julik et al., 2012; Smith et al., 2021). In addition, our muscle maps represent an important contribution to better documentation of precise muscle attachments in large cats; this will be an invaluable help in reconstructing these types of muscle masses and attachments in extinct felid species, such as the American lion (Pantherinae) and sabertooth cats (Machairodontinae).

AUTHOR CONTRIBUTIONS

R Dunn carried out dissection of both limbs, compiled tables and analyzed data, drew muscle maps, and drafted and revised the manuscript. J Meachen initiated the project, assisted in the dissection of both limbs and drafted and revised the manuscript. A Beresheim assisted in the dissection of the left limb, A Gubatina assisted in the dissection of the right limb, and both provided comments on the manuscript. K Bitterman assisted in procurement and storage of the specimen, and compilation of muscle mass data. L. Butaric assisted in procurement of the specimen and provided comments on the manuscript. K. Bejes, S. Kennedy, S. Markham, D. Miller, M. Mrvoljak, L. Roge‐Jones, J. Stumpner, and Cody Walter assisted in the dissection of the left limb, took notes during the dissection, weighed the muscles, and recorded muscle masses.

Dunn, R.H., Beresheim, A., Gubatina, A., Bitterman, K., Butaric, L., Bejes, K. et al. (2022) Muscular anatomy of the forelimb of tiger (Panthera tigris). Journal of Anatomy. 241:119–144. Available from: 10.1111/joa.13636

DATA AVAILABILITY STATEMENT

Muscle mass (in grams) for each muscle was recorded using an Ohaus Scout Pro (model SP401) digital scale and is provided in Table 1.