Body size of Smilodon (Mammalia: Felidae)

Body Size of Smilodon (Mammalia: Felidae)Per Christiansen1* and John M. Harris2

1Department of Vertebrates, Zoological Museum, 2100 Copenhagen O, Denmark2George C. Page Museum, Los Angeles, California

ABSTRACT The body masses of the three large saber-toothed machairodontines, Smilodon gracilis, S. fatalis,and S. populator, were estimated on the basis of 36 oste-ological variables from the appendicular skeleton of ex-tant felids. A new model is introduced that takes thereliability of the predictor equations into account, sincemass estimates are more reliable when computed frommultiple variables per bone. At a body mass range of55–100 kg, S. gracilis was comparable in size to extantjaguars, and S. fatalis was found to be somewhat lighterthan previously assumed, with a body mass range of 160–280 kg, similar to that of the largest extant felid, theSiberian tiger. Smilodon populator was substantiallyheavier and larger than any extant felid, with a body massrange of 220–360 kg. Particularly large specimens of S.populator almost certainly exceeded 400 kg in body mass.The differences from previous estimates are most likelycaused by differences in the databases used for mass es-timation. J. Morphol. 266:369–384, 2005.© 2005 Wiley-Liss, Inc.

KEY WORDS:body mass; felids; Smilodon; allometry;%PE; %SEE

The machairodontine genus Smilodon was a com-mon large predator in the Pleistocene of North andSouth America. Much attention has been given tostudies of its cranial and dental anatomy (e.g., Simp-son, 1941; Kurten, 1954; Emerson and Radinsky,1980; Martin, 1980; Miller, 1984; Akersten, 1985;Van Valkenburgh and Ruff, 1987; Biknevicius et al.,1996). However, the somatic proportions of saber-toothed cats sensu lato often appear to have differedfrom their extant feline counterparts, the formerappearing more stocky and powerful, with propor-tionally shorter and stouter limb bones, especiallythe distal ones (e.g., Merriam and Stock, 1932;Anyonge, 1993; Turner and Anton, 1997).

The best studied of all saber-toothed cats, Smil-odon fatalis, from the late Pleistocene of NorthAmerica, is known from large samples. Akersten(1985) estimated that the Los Angeles County Mu-seum Hancock Collection encompassed in excess of600 skulls, and Miller (1968) estimated that thetotal number of Smilodon specimens in this collec-tion was in excess of 2,100. Far less common is theSouth American S. populator. It is commonly heldthat S. populator differs from S. fatalis by a numberof characters, including larger overall size (Berta,

1985; Kurten and Werdelin, 1990; Turner and An-ton, 1997). The more plesiomorphic S. gracilis ap-pears to have been distinctly smaller than its twocongenerics (Kurten and Werdelin, 1990; Turnerand Anton, 1997). Previous assessments that S. fa-talis was comparable in size to extant lions (e.g.,Kurten and Anderson, 1980; Akersten, 1985) areprobably incorrect, since proportional differences inoverall build made it considerably heavier than asimple evaluation of overall body proportions wouldsuggest (Anyonge, 1993; Turner and Anton, 1997).

This study focuses on body mass estimation in allthree species of Smilodon based on limb bone dimen-sions. Craniodental variables are frequently used inmass estimation analyses but are less satisfactorydue to a possible systematic bias across geologicaltime (e.g., Damuth, 1990) and to theoretical consid-erations concerning the use of nonweight-bearingvariables used to predict body mass (Hylander,1985; see also Fortelius, 1990). Limb bones are di-rectly responsible for support of mass and instresses of locomotion and have been found to showhigh correlations with body mass in terrestrialmammals (e.g., Christiansen, 1999a,b, 2002a). Thedemands on the appendicular skeleton in terms ofsupport of mass and stresses of locomotion in extinctanimals probably did not differ markedly from theconstraints observed in extant animals.

MATERIALS AND METHODSA sample of 23 extant felids from the Zoological Museum in

Copenhagen, representing 10 species, was measured for the pur-pose of this study (Table 1; Appendix A). In order not to confuseintra- and interspecific allometry, averages of each species withmultiple individuals were used for statistical analysis, with theexception of the tiger, which was represented by two distinctsubspecies, the Bengal subspecies (Panthera tigris tigris) and theSiberian subspecies (P. tigris altaica). This exception was madedue to the distinct physical differences between the two subspe-cies (Mazak, 1981; Nowak, 1991; Sunquist and Sunquist, 2002),and the desire to include as many large taxa in the database aspossible (see below).

*Correspondence to: Per Christiansen, Department of Vertebrates,Zoological Museum, Universitetsparken 15, 2100 Copenhagen O,Denmark. E-mail: [email protected]

Published online 18 October 2005 inWiley InterScience (www.interscience.wiley.com)DOI: 10.1002/jmor.10384

JOURNAL OF MORPHOLOGY 266:369–384 (2005)

© 2005 WILEY-LISS, INC.

Nine specimens, represented by individual bones from the Uni-versity of Florida (UF), of Smilodon gracilis were used in thestudy. Five specimens of S. populator from the Zoological Mu-seum (CN) were included in the study. Of those five specimens,only one (CN52) was a partial skeleton, and the rest were indi-vidual bones from different individuals. Additionally, three spec-imens from Museo de La Plata (Buenos Aires, Argentina) wereincluded in the study.

Twenty-five specimens of Smilodon fatalis from the Los An-geles County Museum (Hancock Collection; LACMHC) wereused in the study. Smilodon fatalis is the second commonestmammalian species to be recovered from the Rancho La Breaasphaltic deposits after dire wolves (Canis dirus). Miller (1968)estimated 2,100 individuals in the Hancock Collection based on

cranial material amassed by the Natural History Museum ofLos Angeles County between 1913 and 1915. His estimate didnot include specimens collected by the University of California,Berkeley (1912–1913), the Southern California Academy ofSciences (1909), or the recent excavation of Pit 91 (1969 –present). Due to postentrapment scavenging and trampling ofcarcasses mired in the asphalt seeps, the bones occur in jum-bled masses and in-life association is lost for most individuals(Gilbert, 1927; Shaw and Quinn, 2001). The cranial and post-cranial elements used in this study should all be construed asfrom different individuals. Accordingly, the specimen LAC-MPMS 1-1 of S. fatalis is the only relatively complete skeletonknown (Cox and Jefferson, 1988), and was also included thestudy.

TABLE 1. Average body mass in kilograms, osteological measurements in mm, and areas in mm3 of the included felids

Feliscatus

Leoparduspardalis

Pumaconcolor

Lynxlynx

Acinonyxjubatus

Pantheraleo

Pantheraonca

Pantherapardus

Pantherat. tigris

Pantherat. altaica

Panthera uncia

BM 8.4 10.2 46.4 9.7 39.0 174.3 63.3 58.9 130.0 225.3 37.3HL 104.0 127.8 211.5 170.5 227.5 343.3 226.0 1210.0 297.3 361.0 203.0HC 28.5 34.3 61.0 38.0 53.0 107.8 73.2 70.5 89.0 108.3 60.3HDAP 10.2 11.7 21.1 12.5 17.8 37.7 24.5 24.1 30.2 37.9 19.9HDLM 7.9 10.1 17.8 11.6 15.9 30.8 22.1 20.8 28.5 31.1 18.5HAP 15.3 20.1 37.3 22.0 28.8 61.6 43.1 35.6 57.7 68.3 34.4HLM 14.1 17.5 29.4 21.2 29.8 58.2 39.1 34.6 51.5 58.1 30.6HHA 221.5 326.2 835.9 403.5 788.5 3112.7 1782.3 1190.4 2731.9 3068.7 1037.8HLCAP 8.3 10.4 19.7 11.9 17.8 36.1 22.9 19.6 29.7 36.2 18.7HLCLM 6.5 8.9 16.3 9.2 12.0 31.3 20.0 18.6 23.4 31.9 17.9HLCA 53.5 115.4 321.2 109.8 212.1 1132.4 463.1 366.1 693.8 1154.1 334.9HMCAP 10.6 13.3 25.3 16.0 23.4 43.2 27.3 25.0 39.5 47.2 23.0HMCLM 4.5 6.7 13.7 8.1 11.0 24.7 15.2 12.1 20.0 23.2 11.5HMCA 47.9 89.9 361.2 130.1 257.5 1075.3 413.5 301.9 790.2 1095.2 267.0HCTA 101.4 205.1 682.3 239.8 469.6 2207.6 876.6 668.1 1484.0 2249.3 601.9HAW 13.6 19.2 36.5 20.7 27.7 66.0 44.2 36.9 53.2 66.7 34.7UL 119.0 137.3 217.5 193.5 262.0 369.8 227.7 228.0 315.3 380.0 224.7UDAP 6.2 8.7 14.0 9.0 7.7 27.2 20.8 12.0 24.8 32.4 16.7OL 16.3 23.2 40.3 23.4 39.6 68.5 46.8 45.5 66.6 78.4 41.5FL 119.5 147.3 258.3 202.5 251.0 391.5 252.0 239.8 351.0 416.3 230.0FC 30.0 36.3 58.8 39.0 57.5 103.2 69.2 67.5 86.8 102.3 59.3FDAP 9.5 11.0 18.7 11.9 19.4 32.6 19.6 21.0 26.3 31.6 17.8FDLM 9.7 11.9 18.8 12.7 17.3 33.0 24.4 22.0 29.0 33.7 20.0FLCAP 12.6 15.4 30.0 18.0 26.7 48.5 31.3 30.1 46.0 52.7 27.6FLCLM 6.6 9.3 16.1 10.3 15.8 30.9 19.9 16.0 26.1 31.4 15.9FLCA 83.3 147.6 483.6 185.7 421.2 1507.4 628.1 484.0 1203.1 1653.8 440.9FMCAP 11.8 14.5 27.4 17.5 24.3 47.3 30.9 28.9 40.9 49.9 25.1FMCLM 6.6 8.8 15.1 8.8 13.4 31.0 20.8 17.9 23.6 30.4 18.6FMCA 77.9 130.2 414.1 154.4 326.0 1475.0 646.0 521.2 963.5 1518.6 467.2FCTA 161.2 277.8 897.7 340.1 747.2 2982.3 1274.1 1005.2 2166.7 3172.3 908.1FAW 19.2 26.1 47.4 27.2 42.6 81.1 52.0 47.4 65.9 78.7 45.8TL 123.0 141.5 224.8 209.5 252.0 333.7 214.2 230.3 296.3 356.8 230.5TC 27.5 31.3 54.8 36.0 53.0 94.8 63.2 58.8 84.5 96.2 55.7TDAP 8.2 9.7 17.6 10.7 16.5 29.0 17.6 18.3 26.0 28.8 17.7TDLM 9.4 10.3 17.2 12.2 17.1 31.1 22.7 19.1 28.1 32.6 17.6TAPROX 208.4 338.1 225.2 447.7 907.4 3699.5 1440.9 1175.1 2334.6 3296.0 1024.2TADIST 130.5 647.8 635.3 283.8 570.2 2081.6 828.0 762.6 1402.5 2000.3 647.8

For values of the individual specimens, see Appendix A.Abbreviations: BM, body mass; FAW, distal articular width of femur; FC, least circumference of femoral diaphysis; FCTA, area offemoral distal condyles; FDAP, anteroposterior diameter of femoral diaphysis; FDLM, lateromedial diameter of femoral diaphysis; FL,length of femur; FLCA, area of femoral lateral condyle; FLCAP, anteroposterior diameter of femoral lateral condyle; FLCLM, latero-medial diameter of femoral lateral condyle; FMCA, area of femoral medial condyle; FMCAP, anteroposterior diameter of femoral medialcondyle; FMCLM, lateromedial diameter of femoral medial condyle; HAP, anteroposterior diameter of humeral head; HAW, distalarticular width of humerus; HC, least circumference of humeral diaphysis; HCTA, area of humeral distal condyles; HDAP, anteropos-terior diameter of humeral diaphysis; HDLM, lateromedial diameter of humeral diaphysis; HHA, area of humeral head; HL, length ofhumerus; HLCA, area of humeral lateral condyle; HLCAP, anteroposterior diameter of humeral lateral condyle; HLCLM, lateromedialdiameter of humeral lateral condyle; HLM, lateromedial diameter of humeral head; HMCA, area of humeral medial condyle; HMCAP,anteroposterior diameter of humeral medial condyle; HMCLM, lateromedial diameter of humeral medial condyle; OL, length ofolecranon process; TC, least circumference of tibia; TADIST, area of tibial distal articulating surface; TAPROX, area of tibial proximalarticulating surface; TDAP, anteroposterior diameter of tibial diaphysis; TDLM, lateromedial diameter of tibial diaphysis; TL, length oftibia; UDAP, anteroposterior diameter of ulnar diaphysis; UL, length of ulna.

370 P. CHRISTIANSEN AND J.M. HARRIS

Body masses for all specimens of the extant species had beenrecorded just after death, which is unusual for museum speci-mens, and no literature values were included. One of the includedleopards (CN5661) was previously figured in articles by Chris-tiansen (1999a,b, 2002a) that cited a body mass of 51 kg. This,however, is eviscerated body mass. According to Spector (1956),the viscera of a domestic cat made up 16.81% of total body mass(digestive system: 10.37%; heart: 0.45%; kidneys: 1.07%; liver:3.59%; lungs: 1.04%; spleen: 0.29%). Accordingly, the specimencan be reestimated to have had a total body mass of around 61 kg(Appendix A).

The included long bones were measured from high-resolutionimages taken in direct anterior, posterior, and lateral views. Foreach bone, length and least circumference of the diaphysis weremeasured, along with anteroposterior and lateromedial diame-ters of the diaphysis at the site of least circumference. The latterare often-used variables in analyses of bone strengths in locomo-tion (Alexander, 1983, 1985; Christiansen, 1998). Length valueswere articular lengths except for the ulna, where it was total bonelength (including the olecranon process). Olecranon length isgiven as the distance from the tip of the olecranon process to thecenter of rotation at the midpoint of the proximal trochlea (Fig.1C). All dimensions are given to the nearest 0.1 mm, except leastcircumference and overall length, which are given to the nearest0.5 mm.

The proportions of the humeral head and the proportions of thehumeral and femoral distal condyles (anteroposterior and latero-medial diameters) were measured and their surface areas werecomputed (Fig. 1). Note that, unlike Ruff (1990), total condylar

area is used in this analyses, due to the mechanical loading of thejoints during all limb postures where body mass is supported, notjust the neutral stance. The surface areas of the proximal anddistal articulating facets on the tibia were computed. The proxi-mal articulating surface area was approximated by summing thesurface areas of the two cotyles for the femoral lateral and medialcondyles. The distal surface area was approximated by computingthe area based on the anteroposterior and lateromedial diame-ters, but this is a rather rough approximation, due to the mark-edly three-dimensional nature of this more complex joint (see alsoGodfrey et al., 1991). Total articular width of the distal humerusand femur are given as the lateromedial diameter from the me-dial face of the medial condyle to the lateral face of the lateralcondyle, excluding the epicondyles (Fig. 1E).

Regression lines were fitted to the data by means of Model I(least squares) regression analysis and confidence intervals werecomputed for both the slopes and intercepts. The regression equa-tions were evaluated by computing the F-statistic and correlationcoefficient. However, the correlation coefficient is not an appro-priate method for evaluating the predictive nature of a regressionequation because the correlation coefficient can be high even withhigh residuals (Smith, 1981, 1984), compromising its predictivepower of the equation, especially at the extremes of the database.Thus, the percent prediction error (%PE) and the percent stan-dard error of the estimate (%SEE) (Smith, 1984) were computed.The former is an indication of the average percentile deviation ofthe individual data points from the values predicted by the re-gression equation. The latter value is an indication of the spreadof the estimates. Assuming a normal distribution, 68% of theactual values would fall within � %SEE of the estimated values.

It is advantageous to use more than just one measurement perbone or per specimen because the proportions of many extinctforms often diverged from their modern relatives in some re-spects, e.g., proboscideans (Christiansen, 2004) and many saber-tooth cats (Fig. 2) (Anyonge, 1993; Turner and Anton, 1997).Thus, it is advisable to use a variety of measurements for eachbone, and use a final average as a predictor of body mass. How-ever, the significance of using the %PE and %SEE (or, for thatmatter, the common r) to evaluate the predictive power of anequation will be compromised if a simple final average of theestimated values is used as a best mass estimate because esti-mates based on less reliable equations count equally to estimatesbased on more reliable equations. Accordingly, if multiple mea-surements per bone have been taken it is advantageous to di-rectly incorporate the predictor variables (%PE, %SEE, or even r)into the final average. A bone or a specimen where multiplemeasurements have been taken will yield exactly the same finalaverage with or without incorporation of the predictor variablesinto the final average, if the variables are of the same magnitudefor each regression equation. This is hardly ever the case.

The %SEE and %PE have a direct bearing on the predictivevalues of the equations by formulating the degree of divergence ofthe predicted values. Traditionally, however, the %SEE and %PEhave not been incorporated directly into the estimated valuesand, accordingly, have only been employed in evaluation of thepredictive value of the included regression equations. Incorporat-ing the %PE into the final estimates for a given species, whosemass is predicted from several equations on various data (e.g.,femur length, tibia circumference) or multiple measurements ofthe same bone, should render these more reliable because the%PE gives an estimate of the average percentage of differencebetween any estimated and a true value. For multiple estimates(n) the final average (WM) will thus not be a mere average of theincluded mass values (M1�M2…Mi/n) but rather:

WM � �Mi

1

((Mi/%PEi)* �Mi

1

%PE)*n�2 (1)

where M is the body mass computed by the regression equationsand n is the sample size. In this fashion, the estimated body massvalues from equations with a higher %PE will be down-weighted

Fig. 1. Limb bones of Panthera onca CN 5660. Proximal hu-merus in A, anterior and B, dorsal view, showing measurementsof humeral head lateromedial (LM, in A) and anteroposterior (AP,in B) diameters. The surface area of the humeral head wascomputed as (AP � LM diameters)*depth*1.57, following Ruff(1990), where depth is linear height of condyle. C: Proximal ulna,showing measurement of olecranon length. Distal femur in D,medial view and E, posterior view, showing measurements ofcondyle AP (in D) and LM (in E) diameters, along with articularwidth across condyles (E, long arrow).

371BODY SIZE OF SMILODON

in the final average. Note that each individual value thus com-puted does not constitute a corrected mass value. Only the finalaverage can be considered a corrected mass estimate. The proce-dure works for the %SEE also, or any predictor variable that maybe used (r, SE, SD, etc.).

RESULTS

All forelimb parameters show high correlationswith body mass (Table 2, Fig. 3), and only the ulnahas correlation coefficients �0.95, especially the an-teroposterior diameter of the diaphysis (r �0.9).However, examination of the residuals identifiedoutliers in 12 cases and omission of an outlier spe-cies resulted in often markedly higher correlationcoefficients and lower %SEE’s and %PE’s (Table 2).The lynx (Lynx lynx) was the most frequent outlierspecies, and only in the dimensions of the ulna wasthere another outlier species, the cheetah (Acinonyxjubatus). The cheetah has especially long lowerlimbs for a felid (Gonyea, 1976a; Christiansen,1998), in accordance with its locomotory specializa-tions (Sharp, 1997; Christiansen, 2002b).

The six best forelimb predictors of body mass,based on the %PE’s and %SEE’s, are shown in Fig-ure 3, and in all but one case (anteroposterior diam-eter of the humeral diaphysis) the equations werederived from the restricted dataset, excluding thelynx. Most of the parameters scale nonisometricallywith body mass (Table 2), and the confidence limitsexclude the coefficients for isometry (mass � lengthdimension3.00; mass � area1.5). This indicates that

the animals become progressively stouter with in-creased body size. Only humerus and ulna lengthsscale with isometry (mass � l3.00).

Hindlimb parameters also showed high correla-tions with body size (r � 0.95; Table 3), and againthe omission of an outlier species (Lynx lynx) re-sulted in improved correlations in 11 cases (Table 3,Fig. 4). Femur and tibia lengths and femur leastcircumference scaled isometrically with body mass,but as for the forelimb variables, most other lineardimensions scaled with positive allometry. Surfaceareas also scaled with positive allometry, both thedistal femoral condyles and tibial proximal and dis-tal articulating facets, as could be expected whenhaving to support a nonisometric increase in bodymass (Table 3). The six best hindlimb predictors ofbody mass are shown in Figure 4, and as for theforelimb these were all datasets where an outlierspecies had been omitted.

The body size of Smilodon gracilis varied between56–97 kg for samples where the bone averages hadbeen found by computation of simple averages and51–99 kg for samples where the %PE was taken intoaccount, down-weighting the less reliable estimatedmasses in favor of estimates from equations with alower %PE (Table 4; Appendix B). Weighted samplesnearly unanimously resulted in higher body massestimates. Weighted samples frequently resulted inhigher standard deviations (Table 4; Appendix B).Smilodon gracilis appears to have been comparablein size to extant leopards, pumas, and jaguars (Sun-

Fig. 2. Humeri of Smilodon populator CN52 (A, medial; C, posterior) and 221 kg Panthera tigris altaica CN 5697 in B, medial andD, posterior views. The humerus of P. tigris measures 373 mm in length and 104 mm in diaphyseal least circumference. Thecorresponding values for Smilodon are 369.5 mm and 145 mm.


quist and Sunquist, 2002). Smilodon gracilis ap-pears also to have been distinctly less massive inoverall build than S. populator (see below), sincethere is not as marked a discrepancy between lengthand width estimates (Appendix B). It appears tohave been more similar to S. fatalis (below).

The body size of Smilodon fatalis varied between140–265 kg for simple averages and 157–281 kg for%PE weighted samples (Table 4; Appendix B). As forS. gracilis, weighted samples nearly unanimouslyresulted in higher body mass estimates, indicatingthat a simple average fails to emphasize the propor-tionally heavy build of Smilodon. Weighted samplesalways resulted in higher standard deviations, indi-cating a wider spread (see also Appendix B). Theestimated mass range is similar to the mass range ofextant Siberian tiger males (Sunquist and Sunquist,2002). As found by Anyonge (1993), the lower limbbones (ulna, tibia) generally yielded slightly lowermass values than the upper limb bones (humerus,femur) and bone lengths resulted in markedly lowerestimated masses than bone diameters (AppendixB), in accord with the proportionally stocky andheavy overall build of Smilodon; Fig. 2 (Turner andAnton, 1997). This demonstrates that body mass

should not be computed from just one or two dimen-sions, but rather from a variety of dimensions.

The partial skeleton LACMPMS 1-1 confirmed theabove, with higher body mass values predicted fromproximal (humerus, femur) than distal (tibia) di-mensions, although the ulna yielded comparablevalues to humerus and femur in this specimen (Ta-ble 5). Overall, however, the variation in estimatedbody mass among the various bones, both from in-dividual specimens and from LACMPMS 1-1, wasrather moderate, indicating that Smilodon fataliswas not built markedly different from the extantlarge felids, where the jaguar (Panthera onca) andclouded leopard (Neofelis nebulosa) also have pro-portionally shorter and stockier limbs and a morerobust build than tigers or lions (Gonyea, 1976a,b;Sunquist and Sunquist, 2002). The mass range of S.fatalis from the current study is substantially lowerthan Anyonge’s (1993) estimate of 347–442 kg(based on humeral and femoral dimensions only).

In contrast, Smilodon populator was substantiallylarger than S. fatalis, and fell outside the size rangeof extant felids (Tables 4, 5). Mass averages cor-rected for differences in %PE’s were unanimouslyhigher than uncorrected ones, indicating that mass

TABLE 2. Regression statistics for felid forelimbs, given as Log (body mass) � � � Log (X), where body massis in kilograms and X are either in mm3 (surface areas) or mm (all other variables)

Variable X n ��95% CI � 95% CI F P SE r %SEE %PE

Humerus length 11 �5.036 � 1.601 2.870 � 0.687 89.444 0.000 0.159 0.953 44.15 26.33Humerus lengtha 10 �4.648 � 0.919 2.720 � 0.685 255.308 0.000 0.087 0.985 22.31 14.31Humerus least circ. 11 �2.868 � 0.534 2.531 � 0.298 368.828 0.000 0.081 0.988 20.52 14.35Humerus least circ.a 10 �2.683 � 0.467 2.436 � 0.258 472.886 0.000 0.065 0.992 16.07 12.27Humerus diaphysis AP 11 �1.680 � 0.318 2.526 � 0.240 566.324 0.000 0.066 0.992 16.33 11.37Humerus diaphysis LM 11 �1.487 � 0.494 2.501 � 0.391 209.675 0.000 0.107 0.979 27.81 17.89Humerus diaphysis LMa 10 �1.326 � 0.433 2.389 � 0.338 265.670 0.000 0.086 0.985 21.84 15.15Humerus head AP diameter 11 �1.969 � 0.551 2.339 � 0.354 223.553 0.000 0.103 0.980 26.86 18.46Humerus head AP diametera 10 �1.791 � 0.484 2.236 � 0.307 281.993 0.000 0.083 0.986 21.16 16.00Humerus head LM diameter 11 �1.961 � 0.551 2.396 � 0.364 221.364 0.000 0.104 0.980 27.00 15.70Humerus head LM diametera 10 �1.759 � 0.361 2.279 � 0.236 497.579 0.000 0.063 0.992 15.65 11.74Humerus head surface area 11 �2.049 � 0.591 1.228 � 0.195 202.376 0.000 0.108 0.978 28.34 19.67Humerus lat. condyle AP 11 �1.343 � 0.388 2.332 � 0.299 310.411 0.000 0.088 0.986 22.51 13.85Humerus lat. condyle APa 10 �1.194 � 0.294 2.232 � 0.224 531.041 0.000 0.061 0.993 15.12 11.33Humerus lat. condyle LM 11 �0.932 � 0.487 2.141 � 0.400 146.697 0.000 0.126 0.971 33.74 22.35Humerus lat. condyle area 11 �1.220 � 0.467 1.149 � 0.186 196.451 0.000 0.110 0.978 28.80 20.08Humerus med. condyle AP 11 �1.645 � 0.481 2.370 � 0.344 243.707 0.000 0.099 0.982 25.64 13.74Humerus med. condyle APa 10 �1.463 � 0.311 2.256 � 0.220 561.087 0.000 0.060 0.993 14.68 11.03Humerus med. condyle LM 11 �0.662 � 0.474 2.117 � 0.426 125.505 0.000 0.136 0.966 36.73 22.16Humerus med. condyle LMa 10 �0.511 � 0.365 2.005 � 0.324 203.508 0.000 0.098 0.981 25.20 17.79Humerus med. condyle area 11 �1.137 � 0.482 1.122 � 0.202 174.088 0.000 0.116 0.975 30.75 16.84Humerus med. condyle areaa 10 �0.970 � 0.346 1.065 � 0.136 326.036 0.000 0.078 0.988 19.58 13.22Humerus condyles total area 11 �1.546 � 0.498 1.144 � 0.177 213.953 0.000 0.106 0.980 27.50 18.36Humerus condyles total areaa 10 �1.382 � 0.429 1.093 � 0.151 279.778 0.000 0.084 0.986 21.25 15.33Humerus articular width 11 �1.711 � 0.561 2.186 � 0.363 186.038 0.000 0.113 0.977 29.67 21.01Ulna length 11 �5.294 � 2.021 2.937 � 0.854 60.498 0.000 0.189 0.933 54.53 27.00Ulna lengtha 10 �4.908 � 1.290 2.792 � 0.544 140.241 0.000 0.117 0.973 30.79 18.23Ulna diaphysis AP 11 �0.489 � 0.796 1.846 � 0.676 38.119 0.000 0.230 0.899 69.65 43.24Ulna diaphysis APb 10 �0.828 � 0.654 2.096 � 0.545 78.695 0.000 0.169 0.953 47.62 30.52Olecranon length 11 �2.036 � 0.489 2.290 � 0.301 294.422 0.000 0.090 0.985 23.15 15.29

Values in bold are significantly different from isometry.aExcluding Lynx lynx.bExcluding Acinonyx jubatus.Abbreviations: AP, anteroposterior diameter; circ, diaphysial least circumference; lat, lateral; LM, lateromedial diameter; med,medial.


values above the simple average were computed bymore reliable equations with a lower %PE and thatvery low values were usually not as reliable ashigher ones, as for S. fatalis. Body mass valuesbased on diaphyseal dimensions always greatly ex-ceeded the corresponding values predicted frombone lengths (Appendix B), as also found for S. gra-cilis and S. fatalis, but in S. populator this was evenmore pronounced, indicating that this huge saber-tooth had a particularly massive overall build. Theestimated mass range for S. populator was 205–329kg (simple averages) and 231–358 kg (averages withcorrection for differences in %PE values). This iscomparable to world record Siberian tigers (Wood,1976; but see Sunquist and Sunquist, 2002, for dis-cussion).

From the partial skeleton LACMPMS 1-1 of Smil-odon fatalis and CN52 of S. populator (Appendix B),it is evident that the epipodials, especially the tibia,were markedly shorter in Smilodon than in extantfelids. For instance, the ulna/humerus ratios of sev-eral extant big cats (Appendix A: Panthera leo, 1.08;P. onca, 1.01; P. pardus, 1.09, P. t. tigris, 1.06; andP.t. altaica, 1.05) were unanimously higher than inthe corresponding ratio in LACMPMS 1-1 (0.97) andCN52 (0.94). Among extant felids, these values arecomparable to particularly stocky individuals of theclouded leopards exclusively (range: 0.97–1.04;Christiansen, unpubl.). Similarly, the tibia/femurratio of the extant big cats ranges from 0.85–0.96 (P.leo and P. onca, 0.85; P. pardus, 0.96; P. t. tigris,

0.84; P. t. altaica, 0.86; and Neofelis nebulosa, 0.96),as also found by Gonyea (1976a,b). In S. fatalisLACMPMS 1-1, it is 0.75. In S. populator CN52, thefemoral caput was missing, rendering a length valuetentative, but the femur appears to have beenaround 400 mm, resulting in a ratio of around 0.70,far lower than in any extant felid.

DISCUSSION

It seems strange that Smilodon fatalis was pre-dicted by Anyonge (1993) to have weighed more thanthe estimates for S. populator presented here, giventhat there is wide agreement that the latter is con-siderably the larger of the two species, as also foundin this study. One possible source of error is thatAnyonge’s body mass data were all literature values,whereas the present analysis employs actual liveweights of the skeletal specimens used. However, itseems more likely that the differences are due toAnyonge’s felid sample, although slightly larger(n � 14) than the present one (n � 11), includingmore small species. As found by both Anyonge(1993) and the present study, many osteologicalvariables scale with positive allometry to body mass,indicating that large felids are proportionally morestoutly built than smaller species, as has previouslybeen demonstrated for many other mammals (e.g.,McMahon, 1973, 1975; Biewener, 1989, 1990; Ber-tram and Biewener, 1990; Christiansen, 1999b,2002a).

Fig. 3. Regression plots, with 95% CI about the regression line, of the six best forelimb predictors of body mass in extant Felidaebased on the %SEE and %PE values. In all cases body mass in kg is along the Y-axis. X-axis: A, humeral least circumference; B,humeral diaphysis AP diameter; C, humeral head LM diameter; D, humeral lateral condyle AP diameter; E, humeral medial condyleAP diameter; F, humeral medial condyle surface area. All datasets restricted, except B (see Table 2).


A data sample with many small species wouldintroduce a size-related bias, producing unreliablyhigh body mass estimates for large species, particu-larly if, like Smilodon, they had thick diaphyses andepiphyses (see Christiansen, 1999b, 2002a). Indeed,for variables that overlap between the two studies(humeral and femoral lengths and least circumfer-ences) the slopes in Anyonge (1993) are distinctlysteeper than corresponding slopes in the presentstudy. Using the restricted datasets from thepresent study (Tables 2, 3), the humeral length slope(HL: 2.720 � 0.685) is markedly, albeit nonsignifi-cantly, lower than Anyonge’s value (HL: 3.13) andthe 95% CI of the femur slope (FL: 2.712 � 0.378)excludes the corresponding value from Anyonge (FL:3.20). Anyonge’s least circumferences (HC: 2.65; FC:2.92) are also markedly higher than the correspond-ing ones from the present study (HC: 2.436 � 0.258;FC: 2.780 � 0.292), although within the 95% CI. Itwould appear that a recalculation of Anyonge’s(1993) data with the equations presented here wouldyield considerably lower body mass values for Smil-odon fatalis and the other included taxa in his anal-ysis. Unfortunately, Anyonge did not publish hisdatabase, rendering detailed comparisons impossi-ble.

The included specimens of Smilodon, althoughshowing some size-variation and including largespecimens, are not among the largest known (see,e.g., Merriam and Stock, 1932; Kurten and Werde-lin, 1990). Accordingly, the extreme upper sizeranges of Smilodon are probably not represented inthe current study and, since unusually large speci-mens in extant mammals are always rare, it is un-likely that such specimens will ever be found in thefossil record. Among extant mammals, at least thelarger species, some unusually large specimens sig-nificantly exceed the upper size range of normalindividuals (see, e.g., Wood, 1976; Nowak, 1991;Christiansen, 1999a, 2004), a pattern also present infelids (Sunquist and Sunquist, 2002).

Although males of both lions and tigers exceeding300 kg in body mass have been shot (Wood, 1976;Sunquist and Sunquist, 2002), this is well outsidethe normal size range of even very large males (Fig.5), and adult Siberian tiger males usually rangefrom 220–260 kg (Sunquist and Sunquist, 2002), inaccord with the sizes of the three specimens used inthe present analysis (Table 1; Appendix A). Accord-ingly, unusually large specimens of Smilodon maywell have significantly exceeded the body massranges reported here also. Accordingly, S. gracilis

TABLE 3. Regression statistics for felid hindlimbs, given as Log (body mass) � � � Log (X), where body massis in kilograms and X are either in mm3 (surface areas) or mm (all other variables)

Variable X n �� 95% CI � 95% CI F P SE r %SEE %PE

Femur length 11 �5.175 � 1.749 2.853 � 0.731 77.970 0.000 0.169 0.947 47.55 21.68Femur lengtha 10 �4.794 � 0.907 2.712 � 0.378 273.933 0.000 0.085 0.986 21.49 15.04Femur least circ. 11 �3.304 � 0.521 2.780 � 0.292 465.159 0.000 0.072 0.990 18.13 11.76Femur diaphysis AP 11 �1.892 � 0.410 2.783 � 0.320 385.004 0.000 0.079 0.989 20.05 14.02Femur diaphysis LM 11 �1.874 � 0.468 2.719 � 0.360 293.244 0.000 0.091 0.985 23.21 16.07Femur lat. condyle AP 11 �1.885 � 0.445 2.433 � 0.305 326.393 0.000 0.086 0.986 21.91 13.30Femur lat. condyle APa 10 �1.712 � 0.319 2.327 � 0.215 618.478 0.000 0.057 0.994 13.94 10.95Femur lat. condyle LM 11 �1.112 � 0.447 2.272 � 0.364 199.114 0.000 0.109 0.978 28.59 17.91Femur lat. condyle LMa 10 �0.961 � 0.375 2.166 � 0.301 275.449 0.000 0.084 0.986 21.43 14.84Femur lat. condyle area 11 �1.507 � 0.437 1.182 � 0.162 270.459 0.000 0.094 0.984 24.24 13.53Femur lat. condyle areaa 10 �1.347 � 0.342 1.130 � 0.126 431.478 0.000 0.068 0.991 16.87 11.62Femur med. condyle AP 11 �1.814 � 0.456 2.430 � 0.318 299.399 0.000 0.090 0.985 22.95 10.82Femur med. condyle APa 10 �1.627 � 0.233 2.314 � 0.160 1102.959 0.000 0.043 0.996 10.30 7.42Femur med. condyle LM 11 �0.977 � 0.452 2.180 � 0.371 176.556 0.000 0.116 0.975 30.52 21.06Femur med. condyle area 11 �1.411 � 0.413 1.163 � 0.156 285.909 0.000 0.092 0.985 23.52 16.39Femur condyles total area 11 �1.825 � 0.449 1.176 � 0.151 310.432 0.000 0.088 0.986 22.51 14.35Femur condyles total areaa 10 �1.663 � 0.372 1.128 � 0.124 443.572 0.000 0.067 0.991 16.63 11.02Femur articular width 11 �2.407 � 0.570 2.457 � 0.345 260.619 0.000 0.096 0.983 24.73 16.89Tibia length 11 �5.926 � 2.699 3.211 � 1.144 40.336 0.000 0.224 0.904 67.62 31.68Tibia lengtha 10 �5.621 � 1.573 3.104 � 0.665 115.787 0.000 0.128 0.967 34.15 20.15Tibia least circ. 11 �3.024 � 0.510 2.678 � 0.292 432.065 0.000 0.075 0.990 18.86 11.78Tibia least circ.a 10 �2.823 � 0.374 2.572 � 0.212 784.391 0.000 0.050 0.995 12.31 8.22Tibia diaphysis AP 11 �1.613 � 0.424 2.654 � 0.342 307.779 0.000 0.089 0.986 22.61 15.18Tibia diaphysis LM 11 �1.726 � 0.429 2.668 � 0.337 320.624 0.000 0.087 0.986 22.12 14.00Tibia diaphysis LMa 10 �1.558 � 0.306 2.551 � 0.237 617.653 0.000 0.057 0.994 13.96 8.92Tibia proximal art. area 11 �2.080 � 0.561 1.235 � 0.185 228.310 0.000 0.102 0.981 26.56 17.12Tibia proximal art. areaa 10 �1.884 � 0.431 1.177 � 0.140 374.240 0.000 0.073 0.989 18.20 13.52Tibia distal art. area 11 �1.963 � 0.531 1.284 � 0.188 238.928 0.000 0.100 0.982 25.91 15.87Tibia distal art. area 11 �1.963 � 0.531 1.284 � 0.188 238.928 0.000 0.100 0.982 25.91 15.87Tibia distal art. areaa 10 �1.775 � 0.416 1.225 � 0.145 379.344 0.000 0.072 0.990 18.07 11.62

Values in bold are significantly different from isometry.aExcluding Lynx lynx.Abbreviation, art, articular surface. Other abbreviations as in Table 2.


almost certainly, on occasion, exceeded 130 kg (Fig.5), as in extant jaguars (Nowak, 1991; Sunquist andSunquist, 2002). Particularly large S. fatalis malesmay well have reached a body mass of around 350 kg(Fig. 5), if not more, but unlike Anyonge (1993), wesuggest that this value is best regarded as an outlierto the normal size range of this species.

Smilodon populator is clearly one of the largestfelids ever discovered, rivaled only by the giantPleistocene North American lion (Panthera atrox;see Merriam and Stock, 1932; Anyonge, 1993;Turner and Anton, 1997), and even normal-sizedindividuals are predicted to have rivaled worldrecord Siberian tiger males (see Wood, 1976; Nowak,1991; Sunquist and Sunquist, 2002). There is everyreason to suppose that unusually large S. populatormales exceeded 400 kg in body mass, perhaps sig-nificantly so. This would effectively remove S. popu-lator from the size distribution of extant felids (Fig.5).

The limb ratios (ulna or radius/humerus ratio;tibia/femur ratio) of Smilodon are lower than in anyextant felid, and some have drawn conclusions abouttheir locomotory potential and possible habitatsfrom this (Gonyea, 1976b; Anyonge, 1996; Turnerand Anton, 1997). A coupling with limb ratios andhabitat has been made in extant felids (Gonyea,1976a), but it is not advisable to draw any conclu-sions about peak locomotor performance based onsimple limb ratios, as many parameters other thanin- vs. outlever lengths operate to distinguish fast-

moving from less fast-moving species among extantmammals (Christiansen, 2002b). The heavy, stockybuild of Smilodon does, however, indicate animalsbuilt for power and possible short ambush raids, notsustained speed, and may be more comparable witha forest-type habitat (Gonyea, 1976b).

This is to some extent supported by the habitatsrepresented at the late Pleistocene locality of Ran-cho La Brea. Fossil plants recovered from Rancho LaBrea appear to represent a coastal sage scrub com-munity, with riparian associations alongside the wa-tercourses and chaparral and deep canyon floras inthe local mountains (Stock, 1992). Jefferson (1988)recognized three distinct mammalian associationsin the Late Pleistocene from southern California:horses and ground sloths were common at coastalsites, llamas and pronghorns were prevalent in in-termontane basins, and camels predominated atsites in the Mojave Desert. The Rancho La Breafauna appears to constitute a variation of the coastalassociation; it has fewer camels than the desert as-semblages but more bison and ground sloths thanthe coastal or intermontane biotas. Because of thenearby Santa Monica Mountains, Rancho La Breamay have been more heavily wooded than othercoastal sites; certainly, the majority of the RanchoLa Brea rodents appear to represent forest or chap-arral habitats.

The actual body mass of Smilodon would alsohave had an influence on its estimated prey size andperhaps prey species, since a relationship exists be-

Fig. 4. Regression plots, with 95% CI about the regression line, of the six best hindlimb predictors of body mass in extant Felidaebased on the %SEE and %PE values. In all cases body mass in kg is along the Y-axis. X-axis: A, femoral lateral condyle AP diameter;B, femoral lateral condyle surface area; C, femoral medial condyle AP diameter; D, femoral condyles total surface area; E, tibialdiaphysis least circumference; F, tibial diaphysis LM diameter. All datasets restricted (see Table 3).


TA

BL

E4.

Ave

rage

bod

ym

asse

s�

SD

ofS

mil

odon

grac

ilis

,S

.fa

tali

s,an

dS

.po

pula

tor,

ind

ivid

ual

bon

es

Hu

mer

i:S

mil

odon

grac

ilis

UF

8727

7U

F63

659

UF

8725

0U

F81

727

un

wei

ghte

dw

eigh

ted

un

wei

ghte

dw

eigh

ted

un

wei

ghte

dw

eigh

ted

un

wei

ghte

dw

eigh

ted

74.1

�25

.882

.2�

40.4

71.1

�23

.477

.2�

33.8

86.6

�26

.495

.3�

41.5

82.0

�24

.789

.8�

39.4

Sm

ilod

onfa

tali

sL

AC

MH

C-K

803

LA

CM

HC

-K80

7L

AC

MH

C-K

812

LA

CM

HC

-K81

6L

AC

MH

C-K

817

LA

CM

HC

-K83

1u

nw

eigh

ted

wei

ghte

du

nw

eigh

ted

wei

ghte

du

nw

eigh

ted

wei

ghte

du

nw

eigh

ted

wei

ghte

du

nw

eigh

ted

wei

ghte

du

nw

eigh

ted

wei

ghte

d24

9.4

�53

.526

5.5

�90

.426

3.6

�48

.728

0.0

�85

.522

6.8

�55

.824

6.1

�10

5.8

216.

1�

43.6

232.

5�

82.6

216.

1�

44.3

234.

8�

91.6

213.

4�

43.7

230.

1�

84.3

Sm

ilod

onpo

pula

tor

ML

P10

-1C

N54

CN

11u

nw

eigh

ted

wei

ghte

du

nw

eigh

ted

wei

ghte

du

nw

eigh

ted

wei

ghte

d32

9.2

�77

.235

8.4

�15

0.8

248.

5�

67.7

266.

8�

115.

231

8.6

�77

.034

1.3

�13

0.7

Uln

ae:

Sm

ilod

ongr

acil

isU

F87

279

UF

8532

1u

nw

eigh

ted

wei

ghte

du

nw

eigh

ted

wei

ghte

d75

.4�

30.0

76.7

�21

.995

.1�

37.2

97.1

�28

.7S

mil

odon

fata

lis

LA

CM

HC

-K15

05L

AC

MH

C-K

1506

LA

CM

HC

-K16

10L

AC

MH

C-K

1611

LA

CM

HC

-K15

94L

AC

MH

C-K

1636

un

wei

ghte

dw

eigh

ted

un

wei

ghte

dw

eigh

ted

un

wei

ghte

dw

eigh

ted

un

wei

ghte

dw

eigh

ted

un

wei

ghte

dw

eigh

ted

un

wei

ghte

dw

eigh

ted

171.

9�

33.8

183.

1�

59.0

220.

9�

129.

421

4.6

�66

.918

3.8

�69

.818

8.3

�56

.623

6.1

�88

.224

2.5

�74

.723

0.4

�99

.523

3.4

�72

.121

1.3

�84

.921

4.0

�53

.1S

mil

odon

popu

lato

rM

LP

91-I

I-1-

152

CN

61C

N3

un

wei

ghte

dw

eigh

ted

un

wei

ghte

dw

eigh

ted

un

wei

ghte

dw

eigh

ted

285.

3�

219.

3261

.4�

101.

621

8.4

—25

5.1

—F

emor

a:S

mil

odon

grac

ilis

UF

8265

2u

nw

eigh

ted

wei

ghte

d69

.5�

10.7

72.9

�17

.5S

mil

odon

fata

lis

LA

CM

HC

-K35

73L

AC

MH

C-K

3577

LA

CM

HC

-K35

79L

AC

MH

C-K

3590

LA

CM

HC

-K35

92L

AC

MH

C-K

3604

un

wei

ghte

dw

eigh

ted

un

wei

ghte

dw

eigh

ted

un

wei

ghte

dw

eigh

ted

un

wei

ghte

dw

eigh

ted

un

wei

ghte

dw

eigh

ted

un

wei

ghte

dw

eigh

ted

190.

2�

26.0

205.

4�

72.4

228.

5�

56.1

247.

8�

104.

319

6.2

�27

.321

2.6

�82

.120

7.9

�36

.922

5.7

�91

.526

0.4

�51

.228

1.3

�10

3.8

214.

6�

27.1

228.

3�

67.3

Sm

ilod

onpo

pula

tor

ML

P10

-48

un

wei

ghte

dw

eigh

ted

296.

4�

75.3

319.

8�

122.

7T

ibia

e:S

mil

odon

grac

ilis

UF

8173

1U

F63

500

un

wei

ghte

dw

eigh

ted

un

wei

ghte

dw

eigh

ted

99.3

�19

.097

.4�

68.6

51.1

�9.

456

.2�

19.6

Sm

ilod

onfa

tali

sL

AC

MH

C-K

4898

LA

CM

HC

-K44

16L

AC

MH

C-K

4434

LA

CM

HC

-K44

39L

AC

MH

C-K

4454

LA

CM

HC

-K44

63u

nw

eigh

ted

wei

ghte

du

nw

eigh

ted

wei

ghte

du

nw

eigh

ted

wei

ghte

du

nw

eigh

ted

wei

ghte

du

nw

eigh

ted

wei

ghte

du

nw

eigh

ted

wei

ghte

d18

8.7

�51

.320

9.9

�83

.118

3.0

�46

.920

3.9

�77

.420

1.8

�50

.622

5.0

�89

.618

9.8

�58

.820

7.4

�77

.614

0.1

�34

.815

7.1

�61

.816

0.6

�38

.018

2.2

�79

.1

Wei

ghte

dav

erag

esar

eco

mpu

ted

aspe

rE

quat

ion

1.


tween the size of extant mammalian predators andtheir prey (e.g., Thompson, 1981; Vezina, 1985;Nowak, 1991; Emerson et al., 1994; Sunquist andSunquist, 2002; see also Elliott et al., 1977; Millsand Biggs, 1993). Extant lions and tigers routinelytake large ungulates that can weigh several hun-dred kilos (Nowak, 1991; Sunquist and Sunquist,2002), although lions hunting individually haveboth lower success-rates and usually chase dis-tinctly smaller prey than when hunting in packs(Kruuk and Turner, 1967; Schaller, 1972; Sunquistand Sunquist, 2002). However, Smilodon’s largebody mass and very massive build would suggestthat it, like the tiger (Seidensticker and McDougal,1993; Sunquist and Sunquist, 2002), would have hadboth the body mass and power to single-handedlywrestle large prey to the ground before deliveringthe killing bite (see also Akersten, 1985).

Tigers occasionally kill adult gaurs (Bos gaurus)(Nowak, 1991; Sunquist and Sunquist, 2002), andpack-hunting lions may occasionally take hippos,giraffes, and young elephants (Schaller, 1972; Sun-quist and Sunquist, 2002). Occasionally, some lionprides even attack fully grown elephants (Joubert,

2000), and an intimate proboscidean association hasbeen made with the scimitar-toothed cat Homo-therium (Rawn-Schatzinger, 1992; Marean andEhrhardt, 1995). The notion that Smilodon primar-ily hunted large “pachyderms” was once very popu-lar (Matthew, 1910), but has since been abandoned,and a proboscidean association has not been madefor either the North or South American populations.Some (Gonyea, 1976b; Akersten, 1985) have sug-gested that S. fatalis lived in packs and, if so, itslarge body size may potentially have allowed it tohunt megaherbivores.

The Rancho La Brea assemblage is dominated bycarnivores, but bison, horses, and camels are alsowell represented. Potential large prey items includeproboscideans (Mammuthus columbi, Mammutamericanum), ground sloths (Paramylodon harlani,Nothrotheriops shastensis, Megalonyx jeffersoni),perissodactyls (Equus occidentalis, E. conversidens,Tapirus californicus), and artiodactyls (Camelopshesternus, Antilocapora Americana, Bison antiquus,B. latifrons) (Stock, 1992). A preliminary investiga-tion of carbon and nitrogen isotopes in Rancho LaBrea mammal bone (Coltrain et al., 2004) suggests

TABLE 5. Average body masses of Smilodon fatalis and S. populator, partial skeletons

Smilodon fatalis LACM PMS1-1Humerus Ulna Femur Tibiaunweighted weighted unweighted weighted unweighted weighted unweighted weighted254.5 � 74.8 279.0 � 127.3 236.6 � 50.4 250.4 � 75.3 222.1 � 28.3 239.5 � 80.6 174.2 � 52.5 195.1 � 78.6Smilodon populator CN52Humerus Ulna Femur Tibiaunweighted weighted unweighted weighted unweighted weighted unweighted weighted296.1 � 65.4 316.2 � 118.8 219.9 � 61.2 231.2 � 76.2 230.6 � 43.5 246.8 � 78.6 205.2 � 65.1 238.5 � 129.1

Weighted averages are computed as per Equation 1.

Fig. 5. Size distribution of some extant felids and Smilodon. Dotted lines indicate unusually large specimens. Data for extant felidsfrom Nowak (1991) and Sunquist and Sunquist (2002).


that the larger common carnivores (Canis dirus,Smilodon fatalis, Panthera atrox) were preferen-tially feeding on bison, camel, horses, and ground-sloths, in accordance with the inferences that couldbe made from comparative studies of extant pan-therines.

ACKNOWLEDGMENTS

We thank Susana Bargo and Juan Prevosti(Museo de La Plata), Andrew Currant (Natural His-tory Museum, London), and Christopher A. Shaw(George C. Page Museum, Los Angeles).

LITERATURE CITED

Akersten WA. 1985. Canine function in Smilodon (Mammalia;Felidae; Machairodontinae). Contr Sci Nat Hist Mus Los Ange-les County 356:1–22.

Alexander R McN. 1983. On the massive legs of a moa (Pachyor-nis elephantopus, Dinornithes). J Zool Lond 201:363–376.

Alexander R McN. 1985. Mechanics of posture and gait of somelarge dinosaurs. Zool J Linn Soc 83:1–25.

Anyonge W. 1993. Body mass in large extant and extinct carni-vores. J Zool Lond 231:339–350.

Anyonge W. 1996. Locomotor behaviour in Plio-Pleistocenesabre-tooth cats: a biomechanical analysis. J Zool Lond 228:395– 413.

Berta A. 1985. The status of Smilodon in North and South Amer-ica. Contr Sci Nat Hist Mus Los Angeles County 379:1–15.

Bertram JEA, Biewener AA. 1990. Differential scaling of the longbones in the terrestrial Carnivora and other mammals. J Mor-phol 204:157–169.

Biewener AA. 1989. Mammalian terrestrial locomotion and size.Bioscience 39:776–783.

Biewener AA. 1990. Biomechanics of mammalian terrestrial lo-comotion. Science 250:1097–1103.

Biknevicius AR, van Valkenburgh B, Walker J. 1996. Incisor sizeand shape: implications for feeding behaviors in saber-toothed“cats.” J Vert Paleontol 16:510–521.

Christiansen P. 1998. Strength indicator values of theropod longbones, with comments on limb proportions and cursorial poten-tial. GAIA 15:241–255.

Christiansen P. 1999a. What size were Arctodus simus and Ursusspelaeus (Carnivora: Ursidae)? Ann Zool Fennici 36:93–102.

Christiansen P. 1999b. Scaling of the limb long bones to bodymass in terrestrial mammals. J Morphol 239:167–190.

Christiansen P. 2002a. Mass allometry of the appendicular skel-eton in terrestrial mammals. J Morphol 251:195–209.

Christiansen P. 2002b. Locomotion in terrestrial mammals: theinfluence of body mass, limb length and bone proportions onspeed. Zool J Linn Soc 136:685–714.

Christiansen P. 2004. Body size in proboscideans, with notes onelephant metabolism. Zool J Linn Soc 140:523–549.

Coltrain JB, Harris JM, Cerling TE, Ehleringer JR, Dearing M,Ward J, Allen J. 2004. Trophic level relationships among Ran-cho La Brea fauna and their implications for the paleoecology ofthe late Pleistocene based on bone collagen stable carbon andnitrogen isotope chemistry. Palaeogeogr Palaeoclimat Palaeo-ecol 205:199–219.

Cox SM, Jefferson GT. 1988. The first individual skeleton ofSmilodon from Rancho La Brea. Curr Res Pleistocene 5:66 –67.

Damuth J. 1990. Problems in estimating body masses of archaicungulates using dental measurements. In: Damuth J, MacFad-den BJ, editors. Body size in mammalian paleobiology. Estima-tion and biological implications. Cambridge, UK: CambridgeUniv Press. p 229–253

Elliott JP, Cowan IM, Holling CS. 1977. Prey capture by theAfrican lion. Can J Zool 55:1811–1828.

Emerson SB, Radinsky LB. 1980. Functional analysis of saber-tooth cranial morphology. Paleobiology 6:295–312.

Emerson SB, Greene HW, Charnov EL. 1994. Allometric aspectsof predator-prey interactions. In: Wainwright PC, Reilley SM,editors. Ecological morphology. Integrative organismal biology.Chicago: University of Chicago Press. p 123–139.

Fortelius M. 1990. Problems with using fossil teeth to estimatebody sizes of extinct mammals. In: Damuth J, MacFadden BJ,editors. Body size in mammalian paleobiology. Estimation andbiological implications. Cambridge, UK: Cambridge UniversityPress. p 207–228.

Gilbert JZ. 1927. The bone drift in the tar beds of Rancho La Brea.Bull Southern Calif Acad Sci 26:59–66.

Godfrey L, Sutherland D, Boy D, Gomberg N. 1991. Scaling oflimb joint surface areas in anthropoid primates and other mam-mals. J Zool Lond 223:603–625.

Gonyea WJ. 1976a. Adaptive differences in the body proportionsof large felids. Acta Anat 96:81–96.

Gonyea WJ. 1976b. Behavioral implications of saber-toothed felidmorphology. Paleobiology 2:332–342.

Hylander WL. 1985. Mandibular function and biomechanicalstress and scaling. Am Zool 25:315–330.

Jefferson GT. 1988. Late Pleistocene large mammalian herbi-vores: implications for big game hunters in southern California.In: Zahary RG, editor. Desert ecology 1986: a research sympo-sium. Los Angeles: S Calif Acad Sci S Calif Desert Stud Con-sortium. p 15–39.

Joubert D. 2000. Preying on giants. Natl Geogr 198:84–95.Kruuk H, Turner M. 1967. Comparative notes on predation by

lion, leopard, cheetah and wild dog in the Serengeti area, EastAfrica. Mammalia 31:1–27.

Kurten B. 1954. The Chinese Hipparion fauna. Comment Biol SocSci Fennica 13:1–82.

Kurten B, Anderson E. 1980. Pleistocene mammals of NorthAmerica. New York: Columbia University Press.

Kurten B, Werdelin L. 1990. Relationships between North andSouth American Smilodon. J Vert Paleontol 10:158–169.

Marean CW, Ehrhardt CL. 1995. Paleoanthropological and pale-ontological implications of the taphonomy of a sabertooth’s den.J Hum Evol 29:515–547.

Martin LD. 1980. Functional morphology and the evolution ofcats. Nebraska Acad Sci Trans 8:141–154.

Matthew WD. 1910. The phylogeny of the Felidae. Bull Am MusNat Hist 28:289–318.

Mazak V. 1981. Panthera tigris. Mammal Species 152:1–8.McMahon TA. 1973. Size and shape in biology. Science 179:1201–

1204.McMahon TA. 1975. Allometry and biomechanics: limb bones of

adult ungulates. Am Nat 107:547–563.Merriam JC, Stock C. 1932. The Felidae of Rancho La Brea.

Carnegie Inst Washington Publ 422:1–232.Miller GJ. 1968. On the age distribution of Smilodon californicus

Bovard from Rancho La Brea. Contr Sci Nat Hist Mus LosAngeles County 131:1–17.

Miller GJ. 1984. On the jaw mechanism of Smilodon californicusBovard and some other carnivores. Imperial Valley College MusSoc Occ Pap 7:1–107.

Mills MGL, Biggs HC. 1993. Prey apportionment and relatedecological relationships between large carnivores in Kruger Na-tional Park. In: Dunstone N, Gorman ML, editors. Mammals aspredators. Symp Zool Soc Lond 65:253–268.

Nowak RM. 1991. Walker’s mammals of the world. Baltimore:Johns Hopkins University Press.

Rawn-Schatzinger V. 1992. The scimitar cat Homotherium serumCope. Illinois State Mus Rep Invest 47:1–79.

Ruff C. 1990. Body mass and hindlimb bone cross-sectional andarticular dimensions in anthropoid primates. In: Damuth J,MacFadden BJ, editors. Body size in mammalian paleobiology.Estimation and biological implications. Cambridge, UK: Cam-bridge University Press. p 119–149.


Schaller GB. 1972. The Serengeti lion. Chicago: University ofChicago Press.

Seidensticker J, McDougal C. 1993. Tiger predatory behaviour,ecology and conservation. In: Dunstone N, Gorman ML, editors.Mammals as predators. Symp Zool Soc Lond 65:105–125.

Sharp NCC. 1997. Timed running speed of a cheetah (Acinonyxjubatus). J Zool Lond 241:493–494.

Shaw CA, Quinn JP. 2001. Entrapment events. In: Harris JMeditor. Rancho La Brea: death trap and treasure trove. Terra38:50.

Simpson GG. 1941. The function of saber-like canines in carniv-orous mammals. Am Mus Novitat 1130:1–12.

Smith RJ. 1981. Interpretation of correlations in intraspecific andinterspecific allometry. Growth 45:291–297.

Smith RJ. 1984. Allometric scaling in comparative biology: prob-lems of concept and method. Am J Physiol 15:R152–R160.

Spector WS. 1956. Handbook of biological data. Philadelphia:W.B. Saunders.

Stock C, Harris JM. 1992. Rancho La Brea: a record of Pleistocenelife in California, 7th ed. (revised by J.M. Harris). Nat Hist MusLos Angeles County Sci News 37:1–113.

Sunquist M, Sunquist F. 2002. Wild cats of the world. Chicago:University of Chicago Press.

Thompson PR. 1981. Prey size and evolution in mammalian car-nivores: new tests for an old theory of human evolution, a pilotstudy. Carnivore 5:7–42.

Turner A, Anton M. 1997. The big cats and their fossil relatives.An illustrated guide to their evolution and natural history. NewYork: Columbia University Press.

van Valkenburgh B, Ruff CB. 1987. Canine tooth strength andkilling behavior in large carnivores. J Zool Lond 212:379–397.

Vezina AF. 1985. Empirical relationships between predator andprey size among terrestrial vertebrate predators. Oecologia 67:555–565.

Wood GL. 1976. The Guinness book of animal facts and feats.Middlesex, UK: Guinness Superlatives.


AP

PE

ND

IXA

.B

ody

mas

sin

kilo

gram

s,os

teol

ogic

alm

easu

rem

ents

inm

m,

and

surf

ace

area

sin

mm

3of

the

incl

ud

edsp

ecim

ens

ofex

tan

tfe

lid

s

For

elim

bB

MH

LH

CH

DA

PH

DL

MH

AP

HL

ML

CA

PL

CL

MM

CA

PM

CL

MA

WU

LU

DA

PO

L8.

410

4.0

28.5

10.2

7.9

15.3

14.1

8.3

6.5

10.6

4.5

13.6

119.

06.

216

.3F

elis

catu

sC

N36

29(�

)6.

511

1.0

28.0

8.9

8.9

17.2

15.2

9.2

7.3

11.9

6.1

15.2

115.

58.

918

.6L

eopa

rdu

spa

rdal

isC

N56

65(�

)13

.914

4.5

40.5

14.4

11.3

22.9

19.8

11.5

10.5

14.8

7.2

23.2

159.

08.

527

.8L

eopa

rdu

spa

rdal

isC

N11

11(�

)47

.023

1.5

65.5

22.9

18.8

39.6

30.0

21.4

16.6

27.8

13.9

37.8

230.

512

.143

.2P

um

aco

nco

lor

CN

5663

(�)

45.7

191.

556

.519

.316

.735

.128

.917

.916

.124

.913

.535

.220

4.5

15.9

37.4

Pu

ma

con

colo

rC

N56

64(�

)9.

717

0.5

38.0

12.5

11.6

22.0

21.2

11.9

9.2

16.0

8.1

20.7

193.

59.

023

.4L

ynx

lyn

xC

N12

0(�

)39

.022

7.5

53.0

17.8

15.9

28.8

29.8

17.8

12.0

23.4

11.0

27.7

262.

07.

739

.6A

cin

onyx

juba

tus

CN

4538

(�)

203.

036

6.0

116.

041

.831

.959

.858

.139

.332

.847

.426

.370

.838

6.0

19.5

71.2

Pan

ther

ale

oC

N72

31(�

)15

0.0

318.

094

.032

.327

.557

.655

.033

.532

.339

.321

.662

.434

0.5

28.9

62.9

Pan

ther

ale

oC

N73

64(�

)17

0.0

346.

011

3.5

39.1

33.1

67.3

61.5

35.6

28.8

43.1

26.3

64.9

383.

033

.371

.5P

anth

era

leo

CN

6043

(�)

67.4

231.

078

.526

.723

.244

.043

.925

.021

.329

.415

.748

.123

1.0

15.7

46.2

Pan

ther

aon

caC

N56

60(�

)71

.023

9.0

76.0

25.3

23.1

46.0

38.8

23.6

22.3

28.9

14.2

46.0

235.

523

.550

.4P

anth

era

onca

CN

6221

(�)

51.5

208.

065

.021

.420

.039

.234

.520

.116

.523

.715

.638

.621

6.5

23.1

43.7

Pan

ther

aon

caC

N57

07(�

)61

.020

0.0

62.0

21.5

17.9

31.0

33.0

18.7

17.4

22.9

12.6

34.3

216.

09.

941

.8P

anth

era

pard

us

CN

5661

(�)

56.7

220.

079

.026

.723

.640

.236

.220

.619

.827

.011

.739

.524

0.0

14.1

49.1

Pan

ther

apa

rdu

sC

N56

62(�

)14

5.0

310.

094

.031

.927

.861

.552

.631

.122

.842

.020

.654

.932

0.5

19.2

68.9

Pan

ther

ati

gris

tigr

isC

N56

67(�

)11

5.0

284.

584

.028

.425

.154

.050

.528

.423

.937

.119

.451

.431

0.0

30.3

64.2

Pan

ther

ati

gris

tigr

isC

N56

69(�

)22

1.0

372.

510

4.0

36.4

29.8

68.3

57.2

35.9

32.0

47.2

22.9

66.6

390.

537

.780

.2P

anth

era

tigr

isal

taic

aC

N56

97(�

)22

5.0

360.

510

8.0

37.5

31.3

70.2

61.1

36.0

33.8

46.6

22.5

70.4

374.

038

.280

.3P

anth

era

tigr

isal

taic

aC

N60

49(�

)23

0.0

350.

011

3.0

39.9

32.1

66.4

55.9

36.7

30.0

47.8

24.2

63.2

375.

521

.474

.8P

anth

era

tigr

isal

taic

aC

N56

98(�

)34

.819

7.0

57.5

18.6

17.9

33.2

29.3

18.0

15.8

21.5

10.7

32.6

216.

510

.539

.6P

anth

era

un

cia

CN

5216

(�)

43.1

208.

563

.520

.819

.635

.931

.820

.118

.925

.113

.436

.223

2.0

20.9

44.6

Pan

ther

au

nci

aC

N60

47(�

)34

.020

3.5

60.0

20.2

18.0

33.9

30.8

18.1

18.9

22.5

10.5

35.2

225.

518

.840

.4P

anth

era

un

cia

CN

6046

(�)

Hin

dlim

bB

MF

LF

CF

DA

PF

DL

ML

CA

PL

CL

MM

CA

PM

CL

MA

WT

LT

CT

DA

PT

DL

MT

AD

IST

TA

PR

OX

8.4

119.

530

.09.

59.

712

.66.

611

.86.

619

.212

3.0

27.5

8.2

9.4

208.

4013

0.52

Fel

isca

tus

CN

3629

(�)

6.5

129.

530

.59.

110

.312

.37.

811

.47.

621

.912

8.5

26.5

8.9

8.1

232.

1018

2.28

Leo

pard

us

pard

alis

CN

5665

(�)

13.9

165.

042

.012

.913

.518

.610

.717

.69.

930

.215

4.5

36.0

10.4

12.4

444.

0126

8.08

Leo

pard

us

pard

alis

CN

1111

(�)

47.0

276.

562

.519

.920

.031

.716

.728

.415

.649

.522

5.5

60.5

19.3

19.2

854.

7770

0.16

Pu

ma

con

colo

rC

N56

63(�

)45

.724

0.0

55.0

17.5

17.5

28.3

15.5

26.5

14.6

45.2

224.

049

.015

.915

.110

22.4

157

0.47

Pu

ma

con

colo

rC

N56

64(�

)9.

720

2.5

39.0

11.9

12.7

18.0

10.3

17.5

8.8

27.2

209.

536

.010

.712

.244

7.68

283.

83L

ynx

lyn

xC

N12

0(�

)39

.025

1.0

57.5

19.4

17.3

26.7

15.8

24.3

13.4

42.6

252.

053

.016

.517

.190

7.42

570.

15A

cin

onyx

juba

tus

CN

4538

(�)

203.

040

1.0

109.

035

.233

.855

.333

.451

.933

.387

.134

2.0

102.

029

.534

.942

99.9

924

30.2

1P

anth

era

leo

CN

7231

(�)

150.

037

2.0

95.0

29.9

30.6

42.8

28.9

42.0

28.8

73.2

319.

084

.026

.426

.930

87.7

916

40.5

1P

anth

era

leo

CN

7364

(�)

170.

040

1.5

105.

532

.634

.647

.430

.348

.130

.983

.134

0.0

98.5

31.0

31.5

3710

.81

2174

.12

Pan

ther

ale

oC

N60

43(�

)67

.425

9.5

71.0

19.1

26.0

33.3

20.6

33.6

21.7

53.3

220.

067

.519

.024

.115

40.5

982

5.72

Pan

ther

aon

caC

N56

60(�

)71

.026

5.5

72.5

21.1

25.1

33.1

21.6

32.0

21.5

54.3

221.

564

.017

.223

.615

86.8

095

2.25

Pan

ther

aon

caC

N62

21(�

)51

.523

0.9

64.0

18.7

22.1

27.6

17.5

27.2

19.1

48.4

201.

058

.016

.520

.511

95.1

770

6.09

Pan

ther

aon

caC

N57

07(�

)61

.023

1.0

61.0

18.6

20.2

28.3

14.5

26.6

16.4

43.4

223.

053

.016

.517

.110

64.9

963

8.18

Pan

ther

apa

rdu

sC

N56

61(�

)56

.724

8.5

74.0

23.4

23.7

31.8

17.5

31.2

19.4

51.5

237.

564

.520

.121

.012

85.1

588

7.04

Pan

ther

apa

rdu

sC

N56

62(�

)14

5.0

360.

590

.528

.129

.448

.926

.742

.024

.169

.530

4.5

87.0

25.8

29.7

2483

.17

1438

.97

Pan

ther

ati

gris

tigr

isC

N56

67(�

)11

5.0

341.

583

.024

.428

.543

.125

.639

.823

.062

.428

8.0

82.0

26.1

26.4

2186

.12

1365

.97

Pan

ther

ati

gris

tigr

isC

N56

69(�

)22

1.0

429.

510

2.5

31.2

34.1

51.8

30.4

51.1

30.2

78.5

365.

593

.027

.532

.031

97.5

520

07.4

0P

anth

era

tigr

isal

taic

aC

N56

97(�

)22

5.0

408.

510

2.0

31.4

33.6

53.1

33.0

49.3

32.2

81.2

359.

097

.028

.633

.333

65.0

620

52.6

5P

anth

era

tigr

isal

taic

aC

N60

49(�

)23

0.0

411.

010

2.5

32.1

33.4

53.3

30.7

49.3

28.9

76.3

346.

098

.530

.332

.433

25.4

819

40.9

7P

anth

era

tigr

isal

taic

aC

N56

98(�

)34

.822

5.0

56.5

17.0

18.9

26.4

13.4

23.5

16.9

42.0

226.

054

.017

.217

.189

4.34

678.

29P

anth

era

un

cia

CN

5216

(�)

43.1

234.

563

.019

.121

.128

.315

.926

.818

.547

.823

9.5

58.0

18.3

18.3

1246

.97

708.

14P

anth

era

un

cia

CN

6047

(�)

34.0

230.

558

.517

.419

.928

.218

.425

.120

.347

.522

6.0

55.0

17.5

17.5

931.

2455

6.92

Pan

ther

au

nci

aC

N60

46(�

)

For

elim

bab

brev

iati

ons:

AW

,dis

tal

arti

cula

rw

idth

ofh

um

eru

s;B

M,b

ody

mas

s;H

AP,a

nte

ropo

ster

ior

diam

eter

ofh

um

eral

hea

d;H

C,l

east

circ

um

fere

nce

ofh

um

eral

diap

hys

is;

HD

AP,a

nte

ropo

ster

ior

diam

eter

ofh

um

eral

diap

hys

is;H

DL

M,l

ater

omed

ial

diam

eter

ofh

um

eral

diap

hys

is;H

L,l

engt

hof

hu

mer

us;

HL

M,l

ater

omed

ial

diam

eter

ofh

um

eral

hea

d;L

CA

P,a

nte

ropo

ster

ior

diam

eter

ofh

um

eral

late

ral

con

dyle

;LC

LM

,lat

erom

edia

ldi

amet

erof

hu

mer

alla

tera

lco

ndy

le;M

CA

P,a

nte

ropo

ster

ior

diam

eter

ofh

um

eral

med

ial

con

dyle

;M

CL

M,

late

rom

edia

ldi

amet

erof

hu

mer

alm

edia

lco

ndy

le;

OL

,le

ngt

hof

olec

ran

onpr

oces

s;U

DA

P,

ante

ropo

ster

ior

diam

eter

ofu

lnar

diap

hys

is;

UL

,le

ngt

hof

uln

a.H

indl

imb

abbr

evia

tion

s:A

W,d

ista

lar

ticu

lar

wid

thof

fem

ur;

BM

,bod

ym

ass;

FC

,lea

stci

rcu

mfe

ren

ceof

fem

oral

diap

hys

is;F

DA

P,a

nte

ropo

ster

ior

diam

eter

offe

mor

aldi

aph

ysis

;F

DL

M,l

ater

omed

iald

iam

eter

offe

mor

aldi

aph

ysis

;FL

,len

gth

offe

mu

r;L

CA

P,a

nte

ropo

ster

ior

diam

eter

offe

mor

alla

tera

lcon

dyle

;LC

LM

,lat

erom

edia

ldia

met

erof

fem

oral

late

ral

con

dyle

;M

CA

P,

ante

ropo

ster

ior

diam

eter

offe

mor

alm

edia

lco

ndy

le;

MC

LM

,la

tero

med

ial

diam

eter

offe

mor

alm

edia

lco

ndy

le;

TA

DIS

T,

area

ofti

bial

dist

alar

ticu

lati

ng

surf

ace;

TA

PR

OX,

area

ofti

bial

prox

imal

arti

cula

tin

gsu

rfac

e;T

C,

leas

tci

rcu

mfe

ren

ceof

tibi

a;T

DA

P,

ante

ropo

ster

ior

diam

eter

ofti

bial

diap

hys

is;

TD

LM

,la

tero

med

ial

diam

eter

ofti

bial

diap

hys

is;

TL

,le

ngt

hof

tibi

a.


APPENDIX B (1). Estimated body size of the included specimens of Smilodon gracilis, S. fatalis, and S. populator, individual bones

Humerus:S. gracilis UF87277 S. gracilis UF63659 S. gracilis UF87250 S. gracilis UF81727

%PE Var Uwm Wm Var Uwm Wm Var Uwm Wm Var Uwm WmHL 14.31 238.5 65.8 71.3 233.0 61.8 66.9 259.5 82.8 89.7 260.0 83.2 90.2HC 12.27 83.5 100.2 126.6 72.5 71.1 89.8 84.5 103.1 130.3 85.5 105.8 133.7HDAP 11.37 28.5 98.8 134.8 25.4 73.9 100.8 29.1 104.2 142.1 293 106.0 144.6HDLM 15.15 24.8 101.2 103.6 20.9 67.3 68.9 248 101.2 103.6 25.2 105.2 107.7HAP 16.00 33.7 42.2 40.9 32.2 38.1 36.9 37.8 54.5 52.8 35.1 46.2 44.8HLM 11.74 39.6 76.2 100.7 34.8 56.8 75.0 42.2 88.1 116.4 37.9 69.0 91.1HHA 19.67 2117.5 108.4 85.5 2051.2 104.3 82.2 2160.3 111.1 87.6 1959.8 98.6 77.7LCAP 11.33 27.1 101.0 138.3 25.7 89.7 122.8 27.2 101.8 139.4 24.1 77.7 106.4LCLM 22.35 14.2 34.3 23.8 14.3 34.8 24.1 14.4 35.3 24.5 15.8 43.1 29.9HLCA 20.08 384.8 56.3 43.5 367.5 53.4 41.2 391.7 57.4 44.4 380.8 55.6 43.0MCAP 11.03 33.8 96.9 136.2 33.9 97.5 137.1 35.7 109.6 154.1 35.5 108.2 152.2MCLM 17.79 13.5 56.9 49.6 17.9 100.2 87.4 18.9 111.8 97.4 18.0 101.4 88.4HMCA 13.22 456.3 72.8 85.4 606.8 98.6 115.7 674.7 110.4 129.5 639.0 104.2 122.2HCTA 15.33 841.1 65.3 66.1 974.3 76.7 77.6 1066.4 84.6 85.6 1019.8 80.6 81.5AW 21.01 30.9 35.2 26.0 33.6 42.2 31.2 34.1 43.6 32.2 34.6 45.0 33.2

S. fatalis K803 S. fatalis K807 S. fatalis K812 S. fatalis K816 S. fatalis K817 S. fatalis K831

%PE Var Uwm Wm Var Uwm Wm Var Uwm Wm Var Uwm Wm Var Uwm Wm Var Uwm WmHL 14.31 365.1 209.8 227.4 381.1 235.7 255.5 352.9 191.3 207.3 358.2 199.2 215.9 358.6 199.8 216.5 352.5 190.7 206.7HC 12.27 128.0 281.9 356.4 124.0 261.0 329.9 128.5 284.6 359.8 118.5 233.7 295.4 115.0 217.2 274.6 113.0 208.1 263.1HDAP 11.37 43.6 289.3 394.6 43.1 281.0 383.3 48.1 370.8 505.8 42.4 269.6 367.8 40.4 238.6 325.5 38.7 214.1 292.0HDLM 15.15 37.9 278.8 285.5 35.8 243.3 249.1 33.6 209.1 214.1 32.9 198.9 203.6 32.7 196.0 200.7 33.3 204.7 209.6HAP 16.00 62.8 169.5 164.3 61.7 163.0 158.0 56.3 132.8 128.7 59.2 148.6 144.0 58.0 141.9 137.6 57.8 140.8 136.5HLM 11.74 57.0 174.8 231.0 67.9 260.5 344.2 60.6 201.0 265.6 59.1 189.9 250.8 67.7 258.8 341.9 63.5 223.6 295.4HHA 19.67 5586.2 356.8 281.4 5880.3 380.0 299.7 4551.6 277.5 218.8 4903.3 304.0 239.7 4578.5 279.5 220.4 4722.9 290.4 228.9LCAP 11.33 45.6 322.7 441.8 42.4 274.3 375.5 40.9 253.1 346.5 40.6 249.0 340.9 42.2 271.5 371.6 41.7 264.3 361.8LCLM 22.35 30.2 172.5 119.7 32.1 196.5 136.4 29.8 167.6 116.3 27.8 144.4 100.2 27.6 142.2 98.7 29.1 159.3 110.5HLCA 20.08 1377.1 243.6 188.2 1361.0 240.3 185.6 1218.8 211.7 163.5 1128.7 193.8 149.7 1164.7 200.9 155.2 1213.5 210.7 162.7MCAP 11.03 51.2 247.2 347.7 55.8 300.2 422.1 51.4 249.4 350.7 51.5 250.5 352.3 54.0 278.8 392.0 53.7 275.3 387.1MCLM 17.79 28.6 256.5 223.6 29.7 276.6 241.2 25.5 203.8 177.6 26.9 226.8 197.7 24.9 194.3 169.4 25.9 210.2 183.3HMCA 13.22 1464.3 252.0 295.6 1659.9 288.0 337.9 1310.7 223.9 262.7 1385.4 237.6 278.7 1344.6 230.1 270.0 1390.8 238.5 279.9HCTA 15.33 2841.4 247.0 249.9 3020.9 264.1 267.2 2529.5 217.5 220.1 2514.0 216.1 218.6 2509.3 215.6 218.2 2604.3 224.6 227.2AW 21.01 74.2 238.6 176.2 81.1 289.8 213.9 69.6 207.5 153.2 65.1 179.3 132.3 64.5 175.7 129.7 59.1 145.1 107.1

S. populator MLP10-13 S. populator CN54 S. populator CN11

%PE Var Uwm Wm Var Uwm Wm Var Uwm WmHL 14.31 387.5 246.7 267.3 333.5 164.0 177.7 374.5 224.8 243.6HC 12.27 155.0 449.4 568.1 138.5 341.6 431.9 149.0 408.2 516.0HDAP 11.37 54.0 496.6 677.5 49.6 400.7 546.6 51.5 440.6 601.0HDLM 15.15 44.7 413.6 423.4 38.7 293.1 300.1 43.4 385.4 394.6HAP 16.00 71.2 224.5 217.6 70.5 219.6 212.8 81.4 302.8 293.5HLM 11.74 81.0 389.4 514.5 54.1 155.2 205.1 62.8 218.0 288.1HHA 19.67 4970.2 309.1 243.8 5027.5 313.5 247.2 6474.9 427.8 337.3LCAP 11.33 45.9 327.5 448.3 40.5 247.7 339.0 44.5 305.6 418.3LCLM 22.35 35.1 238.0 165.1 32.3 199.2 138.2 33.1 209.9 145.6HLCA 20.08 1611.1 291.7 225.3 1307.4 229.5 177.3 1472.5 263.1 203.2MCAP 11.03 57.8 325.0 457.0 53.9 277.6 390.4 62.0 380.7 535.4MCLM 17.79 31.4 309.3 269.6 24.1 181.9 158.6 28.8 260.1 226.7HMCA 13.22 1814.9 316.7 371.6 1298.1 221.7 260.1 1785.6 311.3 365.2HCTA 15.33 3426.0 303.1 306.6 2605.5 224.7 227.3 3258.1 286.9 290.2AW 21.01 82.1 297.7 219.8 76.9 258.0 190.5 88.8 353.4 260.9

Ulna:

S. gracilis UF87279 S. gracilis UF85321

%PE Var Uwm Wm Var Uwm WmUL 18.23 228.5 47.6 55.8 248.0 60.0 70.3UDAP 30.52 23.1 107.2 75.0 25.7 134.0 93.7OL 15.29 49.9 71.2 99.4 55.6 91.2 127.4


%PE Var Uwm Wm Var Uwm Wm Var Uwm Wm Var Uwm Wm Var Uwm Wm Var Uwm WmUL 18.23 331.7 134.9 157.9 315.6 117.4 137.4 315.1 116.9 136.8 344.7 150.2 175.8 333.9 137.4 160.9 334.4 138.0 161.5UDAP 30.52 31.2 201.3 140.8 41.5 366.0 256.0 35.0 256.1 179.1 39.3 326.5 228.3 39.8 335.2 234.5 38.0 304.2 212.8OL 15.29 74.7 179.4 250.5 74.7 179.4 250.5 74.5 178.3 249.0 83.5 231.6 323.3 81.4 218.4 305.0 76.9 191.8 267.7

S. populator MLP9I-II-1-152 S. populator CN61 S. populator CN3

%PE Var Uwm Wm Var Uwm Wm Var Uwm WmUL 18.23 350.8 157.7 184.7UDAP 30.52 49.9 538.5 376.7OL 15.29 71.0 159.7 223.0 81.4 218.4 87.1 255.1


APPENDIX B (1). (Continued.)

Femur:S. gracilis UF82652

%PE Var Uwm WmFL 15.04 269.0 62.5 57.9FC 11.76 70.5 68.0 80.5FDAP 14.02 21.1 62.2 61.7FDLM 16.07 23.7 73.1 63.3LCAP 10.95 30.9 56.9 72.4LCLM 14.84 22.7 94.7 88.8FLCA 11.62 701.4 74.0 88.6MCAP 7.42 28.9 56.7 106.4MCLM 21.06 21.1 81.3 53.7FMCA 16.39 609.8 67.3 57.2FCTA 11.02 1311.2 71.4 90.2AW 16.89 52.6 66.3 54.6


%PE Var Uwm Wm Var Uwm Wm Var Uwm Wm Var Uwm Wm Var Uwm Wm Var Uwm WmFL 15.04 370.5 148.8 137.7 390.1 171.1 158.4 379.4 158.7 146.9 384.4 164.4 152.2 423.5 213.8 197.9 401.3 184.8 171.0FC 11.76 108.0 223.3 264.4 118.5 289.0 342.2 101.5 187.9 222.5 111.5 244.0 288.9 122.0 313.4 371.0 113.5 256.4 303.5FDAP 14.02 32.9 214.0 212.5 34.2 238.3 236.7 30.5 173.3 172.1 33.4 223.1 221.6 38.5 331.4 329.1 35.4 262.3 260.5FDLM 16.07 36.0 227.8 197.4 41.1 326.6 283.0 34.2 198.2 171.7 37.7 258.3 223.8 39.1 285.2 247.1 36.7 240.1 208.0LCAP 10.95 54.3 211.3 268.7 60.9 275.9 350.9 57.7 243.4 309.4 55.9 226.1 287.4 67.1 345.8 439.7 52.4 194.5 247.3LCLM 14.84 28.1 150.3 141.0 26.1 128.1 120.2 27.8 146.8 137.8 26.6 133.4 125.2 30.0 173.2 162.5 30.8 183.3 172.0FLCA 11.62 1525.8 178.0 213.3 1589.5 186.4 223.3 1604.1 188.3 225.7 1486.9 172.9 207.1 2031.0 245.9 294.6 1613.9 189.6 227.2MCAP 7.42 49.6 197.9 371.3 55.3 254.5 477.5 52.1 221.7 416.0 54.4 245.0 459.7 54.3 244.0 457.8 50.3 204.4 383.5MCLM 21.06 30.3 178.9 118.3 34.0 229.9 152.0 32.4 207.0 136.9 31.7 197.4 130.5 35.7 255.8 169.1 33.9 228.5 151.1FMCA 16.39 1502.9 192.2 163.3 1880.2 249.4 211.9 1688.0 220.0 186.9 1724.5 225.5 191.6 1938.5 258.4 219.5 1705.2 222.6 189.1FCTA 11.02 3028.7 183.6 232.0 3469.7 214.0 270.4 3292.1 201.7 254.8 3211.4 196.1 247.8 3969.5 249.1 314.7 3319.1 203.6 257.2AW 16.89 78.3 176.2 145.2 78.8 179.0 147.5 83.6 206.9 170.6 83.9 208.8 172.1 83.9 208.8 172.1 83.4 205.7 169.6

S. populator MLP10-48

%PE Var Uwm WmFL 15.04 390.2 171.2 158.5FC 11.76 134.5 411.0 486.6FDAP 14.02 40.4 378.9 376.3FDLM 16.07 45.4 428.1 370.9LCAP 10.95 64.8 318.8 405.4LCLM 14.84 35.0 241.8 226.9FLCA 11.62 2268.0 278.5 333.7MCAP 7.42 57.9 283.0 531.1MCLM 21.06 35.0 244.9 161.9FMCA 16.39 2026.5 272.1 231.2FCTA 11.02 4294.5 272.2 343.9AW 16.89 91.1 255.6 210.7Tibia:

S. gracilis UF81731 S. gracilis UF63500

%PE Var Uwm Wm Var Uwm WmTL 20.15 251.0 67.3 43.2 213.5 40.8 26.2TC 8.22 77.5 108.3 170.5 58.0 51.6 81.2TDAP 15.18 22.9 99.1 84.4 18.3 54.6 46.6TDLM 8.92 26.4 117.1 169.8 18.6 47.9 69.5TAPROX 13.52 1429.1 67.5 64.6TADIST 11.62 1253.0 104.7 116.6 620.5 44.3 49.3


%PE Var Uwm Wm Var Uwm Wm Var Uwm Wm Var Uwm Wm Var Uwm Wm Var Uwm WmTL 20.15 293.2 108.9 69.9 284.8 99.5 63.9 305.2 123.4 79.2 299.7 116.6 74.8 259.9 74.9 48.1 276.4 90.7 58.2TC 8.22 99.5 206.7 325.3 96.0 188.5 296.7 102.5 223.1 351.1 98.0 198.8 312.8 87.5 148.5 233.7 94.5 181.1 284.9TDAP 15.18 33.1 263.4 224.4 31.3 227.1 193.5 33.8 278.4 237.3 34.1 285.1 242.9 28.1 170.6 145.3 29.9 201.1 171.4TDLM 8.92 30.4 167.8 243.3 29.9 160.9 233.3 31.5 183.7 266.4 28.3 139.8 202.7 27.6 131.1 190.2 30.4 167.8 243.3TAPROX 13.52 3705.0 207.2 198.3 3809.6 214.1 204.9 3594.5 200.0 191.3 3699.6 206.9 197.9 3052.1 165.0 157.8 2826.2 150.7 144.2TADIST 11.62 1931.4 177.9 198.0 2190.6 207.6 231.1 2142.3 202.0 224.8 2050.9 191.5 213.2 1685.7 150.6 167.6 1878.8 172.0 191.5

Abbreviations: Var. variable; Uwm. unweighted body mass; Wm. weighted body mass. Other abbreviations as in Appendix A andTable 1.


AP

PE

ND

IXB

(2).

Est

imat

edbo

dy

size

ofth

epa

rtia

lsk

elet

ons

ofS

mil

odon

fata

lis

(LA

CM

PM

S1-

1)an

dS

.po

pula

tor

(CN

52)

S.

fata

lis

LA

CM

PM

S1-

1H

um

eru

sU

lna

Fem

ur

Tib

ia%

PE

Var

Uw

mW

m%

PE

Var

Uw

mW

m%

PE

Var

Uw

mW

m%

PE

Var

Uw

mW

mH

L14

.31

281.

623

6.6

258.

4U

L18

.23

370.

718

4.0

215.

4F

L15

.04

390.

417

1.5

158.

7T

L20

.15

294.

311

0.2

70.1

HC

12.2

713

3.0

309.

539

1.3

UD

AP

30.5

236

.828

4.5

199.

0F

C11

.76

113.

525

6.4

303.

5T

C8.

2295

.018

5.0

288.

4H

DA

P11

.37

48.1

370.

850

5.8

OL

15.2

985

.024

1.2

336.

7F

DA

P14

.02

36.5

285.

728

3.7

TD

AP

15.1

832

.324

6.8

208.

3H

DL

M15

.15

36.7

258.

226

4.4

FD

LM

16.0

735

.822

4.4

194.

4T

DL

M8.

9228

.213

8.5

199.

0H

AP

16.0

063

.717

1.3

166.

1L

CA

P10

.95

56.5

231.

729

4.7

TA

PR

OX

13.5

2—

HL

M11

.74

67.3

255.

333

7.3

LC

LM

14.8

432

.320

3.2

190.

7T

AD

IST

11.6

220

40.0

190.

220

9.8

HH

A19

.67

5789

.337

2.8

294.

0F

LC

A11

.62

1825

.022

2.7

266.

9L

CA

P11

.33

48.1

363.

549

7.7

MC

AP

7.42

52.1

221.

741

6.0

LC

LM

22.3

525

.512

0.1

83.3

MC

LM

21.0

632

.220

4.2

135.

0H

LC

A20

.08

1226

.621

3.3

164.

7F

MC

A16

.39

1677

.621

8.4

185.

6M

CA

P11

.03

54.9

289.

440

6.9

FC

TA

11.0

235

02.6

216.

327

3.3

MC

LM

17.7

924

.318

5.0

161.

3A

W16

.89

83.8

208.

217

1.6

HM

CA

13.2

213

34.1

228.

226

7.7

HC

TA

15.3

325

60.6

220.

522

3.0

AW

21.0

171

.922

2.7

164.

4S

.po

pula

tor

CN

52H

um

eru

sU

lna

Fem

ur

Tib

ia%

PE

Var

Uw

mW

m%

PE

Var

Uw

mW

m%

PE

Var

Uw

mW

m%

PE

Var

Uw

mW

mH

L14

.31

369.

521

6.7

234.

9U

L18

.23

348.

515

4.8

181.

3F

L15

.04

—T

L20

.15

286.

010

0.8

64.7

HC

12.2

714

5.0

382.

048

2.9

UD

AP

30.5

236

.327

6.4

193.

3F

C11

.76

120.

029

9.3

351.

8T

C8.

2210

8.0

255.

340

1.7

HD

AP

11.3

751

.443

8.4

598.

1O

L15

.29

83.0

228.

431

8.9

FD

AP

14.0

235

.827

0.7

266.

9T

DA

P15

.18

32.7

255.

021

7.3

HD

LM

15.1

540

.933

4.5

342.

5F

DL

M16

.07

40.6

315.

927

1.7

TD

LM

8.92

36.2

262.

037

9.9

HA

P16

.00

70.5

219.

621

2.8

LC

AP

10.9

554

.020

8.6

263.

3T

AP

RO

X13

.52

3558

.319

7.6

189.

1H

LM

11.7

460

.119

7.3

260.

6L

CL

M14

.84

31.9

197.

818

4.2

TA

DIS

T11

.62

1774

.316

0.3

178.

5H

HA

19.6

753

51.6

338.

526

6.9

FL

CA

11.6

217

25.4

204.

524

3.2

LC

AP

11.3

345

.832

5.9

446.

1M

CA

P7.

4251

.021

1.0

393.

1L

CL

M22

.35

34.4

227.

915

8.2

MC

LM

21.0

632

.821

2.6

139.

5H

LC

A20

.08

1573

.328

3.9

219.

3F

MC

A16

.39

1673

.821

7.9

183.

7M

CA

P11

.03

53.3

270.

738

0.7

FC

TA

11.0

233

99.2

209.

126

2.3

MC

LM

17.7

931

.931

9.2

278.

3A

W16

.89

80.6

189.

215

4.8

HM

CA

13.2

216

98.6

295.

234

6.3

HC

TA

15.3

332

73.9

288.

429

1.8

AW

21.0

182

.830

3.3

223.

9

Abb

revi

atio

ns:

Var

.va

riab

le;

Uw

m.

un

wei

ghte

dbo

dym

ass;

Wm

.w

eigh

ted

body

mas

s.O

ther

abbr

evia

tion

sas

inA

ppen

dix

Aan

dT

able

1.


Documents

Body size of Smilodon (Mammalia: Felidae)