Endocranial morphology of Labidolemur kayi (Apatemyidae

This article was downloaded by: [University of Toronto Libraries]On: 08 November 2011, At: 07:33Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,37-41 Mortimer Street, London W1T 3JH, UK

Journal of Vertebrate PaleontologyPublication details, including instructions for authors and subscription information:http://www.tandfonline.com/loi/ujvp20

Endocranial morphology of Labidolemur kayi(Apatemyidae, Apatotheria) and its relevance to thestudy of brain evolution in EuarchontogliresMary T. Silcox a , Claire K. Dalmyn b , Andrea Hrenchuk c , Jonathan I. Bloch d , Doug M.Boyer e & Peter Houde fa Department of Social Sciences, University of Toronto Scarborough, 1265 Military Trail,Scarborough, Ontario, M1C 1A4, Canadab Department of Social Anthropology, York University, 4700 Keele St, Toronto, Ontario, M3J1P3, Canadac Department of Biology, University of Winnipeg, 515 Portage Avenue, Winnipeg, Manitoba,R3B 2E9, Canadad Florida Museum of Natural History, University of Florida, P. O. Box 117800, Gainesville,Florida, 32611, U.S.Ae Department of Anthropology and Archaeology, Brooklyn College, 2900 Bedford Avenue,Brooklyn, New York, 11210, U.S.Af Department of Biology, MSC 3AF, New Mexico State University, P.O. Box 30001, Las Cruces,New Mexico, 88003, U.S.A

Available online: 08 Nov 2011

To cite this article: Mary T. Silcox, Claire K. Dalmyn, Andrea Hrenchuk, Jonathan I. Bloch, Doug M. Boyer & Peter Houde(2011): Endocranial morphology of Labidolemur kayi (Apatemyidae, Apatotheria) and its relevance to the study of brainevolution in Euarchontoglires, Journal of Vertebrate Paleontology, 31:6, 1314-1325

To link to this article: http://dx.doi.org/10.1080/02724634.2011.609574

PLEASE SCROLL DOWN FOR ARTICLE

Full terms and conditions of use: http://www.tandfonline.com/page/terms-and-conditions

This article may be used for research, teaching, and private study purposes. Any substantial or systematicreproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form toanyone is expressly forbidden.

The publisher does not give any warranty express or implied or make any representation that the contentswill be complete or accurate or up to date. The accuracy of any instructions, formulae, and drug doses shouldbe independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims,proceedings, demand, or costs or damages whatsoever or howsoever caused arising directly or indirectly inconnection with or arising out of the use of this material.

http://www.tandfonline.com/loi/ujvp20

http://dx.doi.org/10.1080/02724634.2011.609574

http://www.tandfonline.com/page/terms-and-conditions

Journal of Vertebrate Paleontology 31(6):1314–1325, November 2011© 2011 by the Society of Vertebrate Paleontology

ARTICLE

ENDOCRANIAL MORPHOLOGY OF LABIDOLEMUR KAYI (APATEMYIDAE,APATOTHERIA) AND ITS RELEVANCE TO THE STUDY OF BRAIN EVOLUTION

IN EUARCHONTOGLIRES

MARY T. SILCOX,*,1 CLAIRE K. DALMYN,2 ANDREA HRENCHUK,3 JONATHAN I. BLOCH,4 DOUG M. BOYER,5and PETER HOUDE6

1Department of Social Sciences, University of Toronto Scarborough, 1265 Military Trail, Scarborough, Ontario M1C 1A4, Canada,[email protected];

2Department of Social Anthropology, York University, 4700 Keele St, Toronto, Ontario M3J 1P3, Canada, [email protected];3Department of Biology, University of Winnipeg, 515 Portage Avenue, Winnipeg, Manitoba, R3B 2E9, Canada,

[email protected];4Florida Museum of Natural History, University of Florida, P. O. Box 117800, Gainesville, Florida 32611, U.S.A,

[email protected];5Department of Anthropology and Archaeology, Brooklyn College, 2900 Bedford Avenue, Brooklyn, New York 11210, U.S.A,

[email protected];6Department of Biology, MSC 3AF, New Mexico State University, P.O. Box 30001, Las Cruces, New Mexico 88003, U.S.A,

[email protected]

ABSTRACT—Apatemyids are known from the Paleocene and Eocene of Europe, and the Paleocene to Oligocene of NorthAmerica, and may share a special relationship with Euarchontoglires. The only endocast previously described for an apate-myid pertains to Carcinella sigei from the late Eocene of France. Here we present a composite virtual endocast of Labidole-mur kayi derived from high-resolution X-ray computed tomography data, based on partial crania from the late Paleocene(Clarkforkian) and early Eocene (Wasatchian) of the Clarks Fork Basin, Wyoming. Like C. sigei, L. kayi had voluminous,transversely expansive olfactory bulbs, accounting for approximately 12–15% of the endocranial volume. This is similar toCretaceous eutherians, but contrasts with the relatively smaller olfactory bulbs in both the basal gliran Rhombomylus turpa-nensis and in primitive primates (Ignacius graybullianus, Microsyops annectens). Similar to R. turpanensis, I. graybullianus,and the inferred ancestral condition for Microsyops, but unlike C. sigei, L. kayi exhibited exposed caudal colliculi, support-ing the inference that this condition was primitive for Euarchontoglires and Euarchonta. The cranial capacity of L. kayi isestimated at 0.5–0.6 cc, yielding encephalization quotient (EQ) estimates of 0.23–0.28 or 0.42–0.50 depending on the equa-tion used. These values are much lower than estimates for C. sigei, suggesting significant increase occurred in brain size inApatemyidae, perhaps related to elaborations in the family’s specialized manual extractive feeding regime. Similarities withprimitive primates in EQ and the inferred position of the rhinal sulcus may allow for inferences about encephalization andneocorticalization in the common ancestor of Euarchontoglires.

INTRODUCTION

Apatemyidae are an extinct family of mammals from the earlyPaleocene to late Eocene of Europe, and the early Paleoceneto late Oligocene of North America (McKenna and Bell, 1997).Apatemyids have a characteristic suite of dental features, includ-ing enlarged upper and lower central incisors, an I1 exhibitingan unusual ‘can-opener’ morphology, and an enlarged, blade-like, wedge-shaped p2. All apatemyids described from postcra-nial material can be reconstructed as arboreal, and both NorthAmerican and European specimens have been documented thatexhibit elongate second and third manual digits (Koenigswald,1987, 1990; Koenigswald and Schierning, 1987; Bloch et al.,2004a; Koenigswald et al., 2005a, 2005b). These long fingers mayhave been used with the enlarged anterior dentition for forag-ing for wood boring insects, as observed in extant Dactylop-sila spp. and Daubentonia madagascariensis (Koenigswald andSchierning, 1987). The relationships of apatemyids to other mam-mals have long been a source of contention, with suggested affini-ties to “Insectivora” sensu lato (i.e., including a range of primitive

*Corresponding author.

taxa such as palaeoryctids that may not be closely related to liv-ing eulipotyphlans; e.g., Matthew, 1909; Jepsen, 1934; McKenna,1963; Szalay, 1968; West, 1973; Gingerich, 1982; Koenigswald,1990) or to Primates (e.g., Stehlin, 1916; Schlosser, 1918; Heller,1930; Simpson, 1940, 1945; Romer, 1945; MacPhee et al., 1983)being most common in the literature. Silcox et al. (2010b) in-cluded apatemyids for the first time in a broadly sampled cladis-tic analysis, scoring 240 cranial, dental, and postcranial charac-ters for a wide range of living taxa. Their results suggest that ap-atemyids might be related to Euarchontoglires (the supraordinalgrouping including Primates, Dermoptera, Scandentia, Rodentia,and Lagomorpha) but possibly more closely to Glires than to Pri-mates (Fig. 1). Silcox et al. (2010b) followed Scott and Jepsen(1936) in classifying apatemyids in their own order, Apatotheria.

Cranial material has been described for several species ofapatemyids (Matthew, 1921, 1929; Teilhard de Chardin, 1922;Jepsen, 1934; Scott and Jepsen, 1936; West and Atkins, 1970;West, 1973; Koenigswald, 1987, 1990; Koenigswald and Schiern-ing, 1987; Kalthoff et al., 2004; Koenigswald et al., 2005a, 2005b,2009; Silcox et al., 2010b), including Apatemys uintensis, A. bellus,A. chardini, Heterohyus quercyi, H. nanus, Sinclairella dakoten-sis, Carcinella sigei, and Labidolemur kayi. However, with theexception of the last three species, the known material is either

1314

Dow

nloa

ded

by [U

nive

rsity

of T

oron

to L

ibra

ries]

at 0

7:33

08

Nov

embe

r 201

1

SILCOX ET AL.—LABIDOLEMUR ENDOCRANIAL ANATOMY 1315

FIGURE 1. Hypothesis of relationships including Labidolemur kayi,based on an analysis of 240 cranial, dental, and postcranial traits (Silcoxet al., 2010b). Note that although the most parsimonious tree included arelationship between apatemyids and the basal member of Glires, Rhom-bomylus turpanensis, this node was not well supported and a sister-grouprelationship with Euarchonta would require only five additional steps.

too incomplete, or too strongly compressed, to produce an en-docast. Although the cranium of S. dakotensis would likely haveproduced an excellent endocast, it was unfortunately lost by theUnited States Postal Service in 1976 (W. E. Joyce, personal com-munication as cited in Silcox et al., 2010b). The only endocranialreconstruction previously described for an apatemyid pertains toC. sigei, based on a very complete cranium from the “Phospho-rites du Quercy” (late Eocene, France; Koenigswald et al., 2009).This specimen yielded a volume estimate that is likely to be veryaccurate because the cranium is almost entirely uncompressed.However, identification of blood vessels and other surface fea-tures of the endocast is difficult. What’s more, this specimen per-tains to a relatively late occurring and dentally derived memberof the group, so it may not form an accurate representation of theprimitive endocranial morphology for Apatemyidae.

Three crania of the late Paleocene to early Eocene apate-myid L. kayi are known (see Silcox et al., 2010b), including two(USNM 530221, 530208) that are complete enough to yield par-tial endocasts using high-resolution X-ray computed tomographydata (microCT). One of these (USMN 530221) is part of an artic-ulated skeleton that allows for an estimate of body mass (74 g;Armstrong et al., 2011) derived from multiple postcranial ele-ments, an approach that may be preferable to using only den-tal or cranial specimens to generate an estimate (Gingerich andGunnell, 2005). Together these specimens permit a fairly com-plete characterization of the endocranial morphology of L. kayi,and for the development of a set of estimates of volume for com-posite endocasts produced using different assumption sets.

The aims of this study are twofold. First, we seek to considerevolutionary change in apatemyid endocranial anatomy and rel-ative volume through comparisons between L. kayi, the mostprimitive and oldest apatemyid for which an endocast is known,and the later more dentally derived C. sigei. Second, we considerthe potential relevance of L. kayi to reconstructing primitive en-

docranial morphology for Apatemyidae, Euarchontoglires, andPrimates.

Institutional Abbreviations—UM, University of Michigan Mu-seum of Paleontology, Ann Arbor, Michigan; USNM, Depart-ment of Paleobiology, National Museum of Natural History,Smithsonian Institution, Washington, D.C.

MATERIALS AND METHODS

Two specimens were analyzed in detail in this study (see Silcoxet al., 2010b, for detailed cranial descriptions): (1) USNM 530221,a nearly complete, articulated skeleton with a skull and dentariesfrom UM locality SC-26 (Houde site 14), Clarks Fork Basin,northeastern Wyoming, Willwood Formation, early Wasatchian,lowermost Cardiolophus radinskyi interval zone, early Eocene(between 54.92 and 54.70 Ma); and (2) USNM 530208, an asso-ciated rostrum and basicranium from UM locality SC-62 (BlockZ), Clarks Fork Basin, northwestern Wyoming, lower WillwoodFormation, middle Clarkforkian, uppermost Plesiadapis cookeirange zone, late Paleocene (between 55.68 and 55.36 Ma; see Gin-gerich, 2003, for age model). Although the skull of USNM 530221is somewhat crushed, it preserves the rostral portion of the en-docranial cavity fairly completely. While USNM 530208 is muchless complete, its caudal portion is very well preserved, allowingfor fine resolution of endocranial details.

Both specimens were prepared by P.H. by acid-etching themout of limestone. The limestone was dissolved by submersionin dilute, buffered 7% acetic acid for 2 days at a time. To pro-tect fossils from acid-etching, a thin coating of PVA (polyvinylacetate) was applied to the surface of all exposed bones. Thismode of preparation is responsible in part for the fine sur-face detail on the virtual endocast of USNM 530208, becausethis specimen was almost entirely free of matrix. Comparisonswere made to hemi-endocasts of Tupaia, Sciurus, and Saimiri(Carolina Biological Supply, Bobbitt Laboratory) housed in theUniversity of Winnipeg Anthropology Museum. Comparisonswere also made to a reconstruction of the endocast of P. cookeimade by P. D. Gingerich (see Gingerich and Gunnell, 2005)and to published images and/or data from Moodie (1922), LeGros Clark (1924, 1926, 1932, 1945), Wood (1937), Gazin (1965),Radinsky (1967, 1970, 1975, 1977, 1978), Szalay (1969), Stephanet al. (1970, 1981), Szalay and Berzi (1973), Gingerich and Mar-tin (1981), Gurche (1982), Novacek (1982), Pirlot and Kamiya(1982), Kielan-Jaworowska (1984), Martin (1990), Simons andRasmussen (1996), Meng et al. (2003), Bush et al. (2004), Dozoet al. (2004), Gingerich and Gunnell (2005), Macrini et al. (2007),Sears et al. (2008), Koenigswald et al. (2009) and Silcox et al.(2009, 2010a).

Both USNM 530221 and 530208 were scanned with theOMNI-X HD-600 Industrial Scanner at the Center for Quantita-tive Imaging (CQI), Pennsylvania State University. For USNM530221, only the skull (and bones in its vicinity) was scanned.USNM 530208 is broken into rostral and caudal pieces (seeSilcox et al., 2010b:fig. 3). Only the caudal portion wasscanned—the more rostral portion is too crushed to permit a re-construction of the endocast. These specimens were immobilizedin floral foam (i.e., ‘oasis’) and scanned in volume mode, in which21 individual two-dimensional slices were created for each rota-tion. The axial fan angle was small enough to assume parallelbeam reconstructions. Each rotation consisted of 2400 views ofthe object spanning 360 degrees. The post-acquisition reconstruc-tion process included all 2400 views, and each individual slice wasstored as a 1024 × 1024 matrix of 16 bit integers in tiff format.For USNM 530221, the reconstructed pixel size was 0.053 mmand the interslice distance was 0.058 mm; the data set included697 images. For USNM 530208, the reconstructed pixel size was

Dow

nloa

ded

by [U

nive

rsity

of T

oron

to L

ibra

ries]

at 0

7:33

08

Nov

embe

r 201

1

1316 JOURNAL OF VERTEBRATE PALEONTOLOGY, VOL. 31, NO. 6, 2011

FIGURE 2. Unmodified dorsal views of virtual endocasts of Labidolemur kayi based on A, USNM 530208 and D, USNM 530221, and compositeendocasts based on the B, Rigid Warp and C, Bookstein Warp algorithms in Amira 3.1.1. Scale bar equals 5 mm.

0.033 mm, the interslice distance was 0.036 mm, and the data setconsisted of 361 images.

The images were initially cropped using crop16bit, a DOS pro-gram written by Nathan Jeffrey (University of Liverpool), toremove blank areas. The slices containing the braincase weremanually segmented in ImageJ (Rasband, 1997–2008). In otherwords, the area that the brain would have occupied was filled withpure white to allow it to be selected independently of the speci-men. When pieces of bone were missing, a straight line was drawnbetween the preserved edges. The segmented slices were loadedinto ImageJ as an Image Sequence; the stack was converted to8-bit, and was then loaded in Amira 3.1.1 (Visage Imaging). A la-belfield module was attached, and using the Image SegmentationEditor each slice of the endocast was labeled. A surfacegen mod-ule was attached to the labels file and used to produce the surfacerendering of the endocasts (Fig. 2A, D).

For USNM 530221, a portion of the skull is out of anatomicalposition on the left side; this was shifted into place using manualregistration via landmarks. Two approaches were taken in Amira3.1.1 to form a composite reconstruction. Both depend on select-ing landmarks in the two surface reconstructions, and using thoseto allow the surfaces to be registered to one another. In the rigidwarp approach, the landmarks are aligned but neither surface isaltered (Fig. 2B). This has the advantage that it does not modifythe original data sets, and works well when the two specimens be-ing combined are of similar size and shape. However, in this case,USNM 530221 is clearly from a smaller individual than USNM530208, so the composite generated using this approach is mis-proportioned, i.e., the rostral portion is too small, and the caudalportion too large. The second option, called a Bookstein warp, al-ters one surface reconstruction to fit with the second. In this case,USNM 530208 was resized and warped to fit with USNM 530221.The latter specimen was used as the reference because a reliablebody mass estimate based on multiple postcranial elements waspossible. The resulting composite (Fig. 2C) does not appear to beobviously mis-proportioned, and is probably a closer representa-tion of the original size of the endocast of USNM 530221 thanthe composites produced using the rigid warp algorithm. The dis-advantage to the technique is that details visible in the unalteredvirtual endocast of USNM 530208 are lost in the process of com-bining the two reconstructions. For this reason, the composite isused to estimate volume (Table 1) and measurements such astotal length (Table 2), whereas detailed morphology is assessedfrom the separate endocranial data sets. Volume estimates basedon all assumption sets are provided in Table 1.

Linear measurements for the composite virtual endocast listedin Table 2 were done in Amira 3.1.1, and in all cases representmaximum values for a line fitted to the external surface of the en-docast. For example, the maximum width represents the length ofthe longest line that could be fitted to the endocast perpendicularto the main rostrocaudal axis of the endocast, as defined by thesuperior sagittal sinus. Because these represent maximum values,particular landmarks were not employed in taking these mea-surements. All graphs were produced using SPSS version 12.0 forWindows.

DESCRIPTION AND COMPARISONS

Olfactory Bulbs

Morphology of the olfactory bulbs is only observable in thevirtual endocast derived from USNM 530221 (Fig. 3C, E, F).The olfactory bulbs are offset from the cerebrum by a circularfissure, but are nearly as broad caudally as the rostral end ofthe cerebrum, reflecting their relatively large size (Fig. 3C,E). The olfactory bulbs make up 12–15% of the total volumeof the endocast, depending on the assumptions used in formingthe composite (Table 1), and almost a third of the length ofthe composite produced using the Bookstein warp (Fig. 2C;Table 2). The volume of the olfactory bulbs relative to thatof the overall endocast of Labidolemur kayi is in the range of‘basal insectivores,’ small mammals with relatively primitivecerebral patterns (Stephan, 1972), Cretaceous eutherians such asAsioryctes, and the Oligocene leptictid Leptictis (Fig. 4). In con-trast, all euarchontoglirans included in this study had relativelysmaller olfactory bulbs. For example, although plesiadapiformshave relatively large olfactory bulbs for primates, the maximumpercentage volume observed (5.53% for Ignacius graybullianus;Silcox et al., 2009, 2010a) is markedly lower than even the lowestestimate for L. kayi. Unfortunately quantitative data for theolfactory bulbs have not been published for two of the most rele-vant fossil taxa known from endocasts: the apatemyid Carcinellasigei (Koenigswald et al., 2009) and the basal member of GliresRhombomylus turpanensis (Meng et al., 2003). Nonetheless, it isclear that this region of the brain was also relatively voluminousin the former, based on the olfactory bulbs’ broad bases, and thefact that they account for more than a quarter of the total lengthof the endocast (Koenigswald et al., 2009:fig. 12). In R. turpanen-sis, on the other hand, the olfactory bulbs are relatively narrower,lacking the transverse expansion seen in the apatemyids (Meng

Dow

nloa

ded

by [U

nive

rsity

of T

oron

to L

ibra

ries]

at 0

7:33

08

Nov

embe

r 201

1


FIGURE 3. Virtual endocasts of Labidolemur kayi based on USNM 530208 (A, B, D) and USNM 530221 (C, E, F) in A, E, dorsal, B, F, ventral, andC, D, left lateral views. Scale bar equals 5 mm. (Color figure available online).

et al., 2003:fig. 50). They constitute only approximately a fifth ofthe total length of the endocast in R. turpanensis, a proportionsimilar to that observed for the plesiadapiforms I. graybul-lianus and Microsyops annectens (Silcox et al., 2009, 2010a).These observations suggest that voluminous olfactory bulbswere characteristic of apatemyids, who were distinct in this

feature from euarchontoglirans for which this morphology isknown.Cerebrum

The presence of a well-demarcated circular fissure sug-gests that the cerebrum would not have overlapped onto the

Dow

nloa

ded

by [U

nive

rsity

of T

oron

to L

ibra

ries]

at 0

7:33

08

Nov

embe

r 201

1


TABLE 1. Endocast and olfactory bulb volume estimates for Labidolemur kayi under different assumptions sets.

Assumption set∗Endocranial

volume (mm3) EQ (Jerison) EQ (Eisenberg)Olfactory bulbvolume (mm3)

Olfactory bulb volume as apercentage of the total

endocranial volume

1. Rigid warp without altering USMN 530221 607.785 0.28 0.50 74.05 12.182. Rigid warp, with a portion of USNM

530221 shifted into anatomical position(Fig. 3B)

605.522 0.28 0.50 74.05 12.23

3. Composite with USNM 530208 resized butnot warped, portion of USNM 530221shifted into anatomical position

501.876 0.23 0.42 74.05 14.75

4. Bookstein warp of USNM 530221 andresized USNM 530208, with a portion ofthe former shifted into anatomical position(Fig. 3C)

524.127 0.24 0.43 68.22 13.02

∗Assumption set 1 involves the least manipulation of the data, with the two endocasts being simply overlain on a series of landmarks. For set 2 aportion of the left side of the cranium that was out of place was moved into anatomical position, but neither reconstruction was resized. In set 3,USNM 530208 was resized down to fit better with USNM 530221, but no warping was imposed, whereas for set 4 a Bookstein warp was imposed toproduce the best possible fit of the two surface reconstructions to one another. Encephalization Quotients (EQ) were calculated based on equationsin Jerison, 1973 and Eisenberg, 1981 as indicated. An estimated body mass of 74 g (based on postcranial measurements; Armstrong et al., 2011) wasused for calculating EQ.

olfactory bulbs in L. kayi (Fig. 3C, E). This is also true of otherearly Tertiary euarchontoglirans, including even primitive eupri-mates (adapoids and omomyoids) that differ from extant eupri-mates in which the cerebrum does overlap onto the olfactorybulbs (Gurche, 1982; Meng et al., 2003; Silcox et al., 2009, 2010a).Both virtual endocasts of L. kayi show that the cerebrum had noneocortical sulci (Fig. 3A, E). Amongst modern groups, brainsof less than 5 g typically do not exhibit these features (Macriniet al., 2007). Because the brain of L. kayi would have weighedsignificantly less than 5 g (see below), it would not be expectedto have exhibited neocortical sulci. Notable, however, is the lackof a distinct sylvian sulcus (Fig. 3C), a feature of most livingprimates (except tarsiers and Daubentonia; Preuss, 2009) alsopresent in most fossil euprimates (Le Gros Clark, 1945; Radin-sky, 1967, 1970; Gurche, 1982; Martin, 1990; except Smilodectes;Gazin, 1965; see discussion in Silcox et al., 2010a), but missingin other living and fossil euarchontoglirans, including C. sigei(see Koenigswald et al., 2009:pl. 2), R. turpanensis (see Menget al., 2003:fig. 51), and all plesiadapiforms known from the rele-vant portion of the endocast (Silcox et al., 2009, 2010a). In con-trast to all early Tertiary fossil euprimates known from endocasts(including Smilodectes; Gazin, 1965; Silcox et al., 2010a), but incommon with plesiadapiforms, C. sigei, R. turpanensis, and theMiocene rodent Hypsosteiromys sp., L. kayi also lacks a well-developed temporal pole (Fig. 3C, D; Meng et al., 2003; Dozoet al., 2004; Koenigswald et al., 2009; Silcox et al., 2009, 2010a).

TABLE 2. Measurements from the composite endocast for Labidole-mur kayi, produced using a Bookstein Warp (assumption set 4; Fig. 2C).

Total length 17.51Maximum width 10.47Maximum depth 8.41Olfactory bulb length 5.42Olfactory bulb length/total length 0.31Ratio of olfactory bulb length to the length of rest of the brain 1:2.2Olfactory bulb width 3.25Endocast width/length ratio 0.60Endocast height/length ratio 0.48Endocast height/width ratio 0.80Right optic nerve cross-sectional area 0.48Left optic nerve cross-sectional area 0.36Average optic nerve cross-sectional area 0.42

Measurements in mm or mm2.

A clearly defined orbitotemporal canal (sometimes referred toas a ‘sinus canal,’ but see discussion in Wible and Gaudin, 2004;Wible, 2008) can be identified in USNM 530208 (Fig. 3D). Thisstructure has a close relationship with the rhinal sulcus in extantlemuriforms (Martin, 1990), and has been interpreted as a land-mark for the sulcus in the endocasts of primitive primates (Gazin,1965; Gurche, 1982; Martin, 1990; Silcox et al., 2009, 2010a), andin the Oligocene leptictid Leptictis (Novacek, 1982). The rhinalsulcus (or fissure) is the external marker of the division betweenthe neo- or iso-cortex and the paleocortex (Jerison, 1973, 1991),making its position a coarse indicator of the degree of neocor-ticalization. The orbitotemporal canal is located between a halfand two thirds of the way down the lateral aspect of the cere-brum in L. kayi (Fig. 3D), a position broadly similar to that ob-served in I. graybullianus and M. annectens (Silcox et al., 2009,2010a). This contrasts with the more dorsal position suggestedfor the rhinal fissure in Plesiadapis cookei by Gingerich and Gun-nell (2005). However, the indentation that these authors identifyas the rhinal fissure may in fact represent a lateral sulcus, basedon the presence of a sulcus identified as such in a similar positionin M. annectens (Silcox et al., 2010a). In contrast to that of L. kayiand plesiadapiforms, when it can be observed the orbitotemporalcanal or rhinal sulcus is generally more ventrally located in fossiland living euprimates (e.g., see Gurche 1982:fig. 5). In many liv-ing primates the rhinal sulcus is no longer identifiable, because ofovergrowth of the neocortex (Martin, 1990).

No trace of either the orbitotemporal canal or a separate sulcusis identifiable in the endocasts of either R. turpanensis or C. sigei.In the latter case, this is likely a product of the difficulty of visual-izing structures on the surface of the reconstructed endocast (seeKoenigswald et al., 2009:fig. 12). Meng et al. (2003) interpretedthe absence of the rhinal sulcus in R. turpanensis as a primitivefeature. However, Macrini et al. (2007) note that the rhinal sulcusmay be absent on the endocast of mammals with small brains,even if the feature is present on the brain, and Silcox et al. (2010a)suggested that its absence on the endocast of R. turpanensis waslikely a product of the poor quality of preservation of theendocast surface. Dozo et al. (2004) similarly suggested that poorpreservation could explain the absence of this feature in an endo-cast of Hypsoteiromys. The identification of the rhinal sulcus onthe endocasts of Cretaceous eutherians by Kielan-Jaworowska(1984) would make it surprising if it were truly absent in R.turpanensis, although Macrini et al. (2007:table 3, character 6)apparently disagree with this identification. In any case, it is

Dow

nloa

ded

by [U

nive

rsity

of T

oron

to L

ibra

ries]

at 0

7:33

08

Nov

embe

r 201

1


FIGURE 4. Bivariate plots of ln olfactory bulb volume versus A, ln intracranial volume, and B, ln body mass for an array of living and fossil mammals.Range of values for Labidolemur kayi (indicated by the arrow) in A represents different assumption sets for the estimation of intracranial volume(see Table 1). Range of values presented for Ignacius graybullianus and Microsyops annectens in B reflect varying body mass estimates, includingconfidence intervals (Silcox et al., 2009, 2010a). Data from Stephan et al. (1970, 1981), Gurche (1982), Novacek (1982), Pirlot and Kamiya (1982),Kielan-Jaworowska (1984), Martin (1990), Silcox et al. (2009, 2010a), and the current study (see Supplementary Table 1). The lines plotted representleast squares regression lines. Designation of taxa as ‘basal’ vs. ‘progressive’ insectivores follows Stephan (1972), who indicated that the ‘basal’ formshad relatively primitive cerebral patterns, whereas the ‘progressive’ forms “reveal distinct marks of higher development” (p. 156).

Dow

nloa

ded

by [U

nive

rsity

of T

oron

to L

ibra

ries]

at 0

7:33

08

Nov

embe

r 201

1


FIGURE 5. High-resolution X-ray computed tomography slices of USNM 530208. Slices portrayed were taken in the coronal plane at the levelsindicated with horizontal lines on the ventral view of USNM 530208. In the slices ventral is towards the top of the image. The microCT data setcomprises 361 slices collected moving rostrally (i.e., slice no. 1 is at the caudal extent of the cranium). (A) Slice no. 289. The white arrow indicatesthe foramen ovale. (B) Slice no. 155. The white arrow indicates the internal auditory meatus. (C) Slice no. 99. The white arrow indicates the jugularforamen. (D) Slice no. 77. The white arrow indicates the hypoglossal foramen. (Color figure available online).

not possible to compare the location of the orbitotemporalcanal in L. kayi to any corresponding feature of the endocast ofR. turpanensis.

Midbrain

The caudal end of the cerebrum is well preserved in USNM530208 and ends too far rostrally to cover the caudal colliculi (Fig.3A). Exposure of these midbrain structures is also observed inmany early Tertiary euarchontoglirans (i.e., P. cookei, I. graybul-lianus, Microsyops cf. elegans, one specimen of M. annectens, R.turpanensis; Ischyromys sp.; Wood, 1937; Meng et al., 2003; Sil-cox et al., 2009, 2010a) but not in C. sigei, a different specimen ofM. annectens, Megadelphus lundeliusi, the fossil rodent describedby Moodie (1922; species unknown), or any known euprimate(Szalay, 1969; Koenigswald et al., 2009; Silcox et al., 2009, 2010a).This suggests that exposed caudal colliculi might have been prim-itive for Euarchontoglires (also supported by their exposure inCretaceous eutherians including Asioryctes; Kielan-Jaworowska,1984), with expansion of the cerebrum caudally to cover themoccurring independently within Apatemyidae, Glires, Microsy-opidae, and Euprimates. Exposure of the midbrain also occursin living dermopterans, but this has been argued to be a productof an expansion of the colliculi, rather than representing a prim-itive feature (Gingerich and Gunnell, 2005); indeed, it has beensuggested that midbrain exposure can appear as a derived fea-ture, rather than a primitive one (Edinger, 1964), so caution isnecessary in interpreting its significance.

Cerebellum

Morphology of the cerebellum is well preserved in the virtualendocast of USNM 530208. In this specimen, it is possible to iden-tify the vermis and left lateral lobe of the cerebellum (Fig. 3A),separated by a paramedian fissure, as is typical of therians gener-ally (Macrini et al., 2007). The portion of the cranium that wouldhave housed the right lateral lobe and paramedian fissure is miss-ing. Although there is no established method for estimating thesize of the cerebellum from an endocast independent of the massof the brain, in taxa in which the brain is not significantly flexed,some coarse indication of its size can be gleaned from the per-centage that it contributes to the total length of the endocast. Sil-cox et al. (2010a) estimated that the cerebellum constituted abouta quarter of the total length in M. annectens, R. turpanensis, I.graybullianus, and Adapis parisiensis. In contrast, the cerebellumis much shorter in C. sigei, constituting approximately 13% of thetotal length of the endocast, and this region also appears to havebeen quite short in L. kayi. Although there is some damage to therelevant part of USNM 530208, it is clear that on the left side thecaudal extent of the occipital bone is preserved (Fig. 6), implyingthat in the intact cranium there would not have been space cau-dally for a much more extensive cerebellum. It is perhaps surpris-ing that a region of the brain generally understood to be relatedto motor control would be so short in a family thought to haveengaged in complex extractive foraging behaviors (Koenigswaldand Schierning, 1987).

The fissura prima cannot be identified on the surface of thecerebellum in either virtual endocast of L. kayi. As observed in

Dow

nloa

ded

by [U

nive

rsity

of T

oron

to L

ibra

ries]

at 0

7:33

08

Nov

embe

r 201

1


FIGURE 6. Virtual endocast of Labidolemur kayi (USNM 530208) in-side a translucent rendering of the cranium in A, ventral and B, lateralviews. Scale bar equals 5 mm.

other fossil euarchontoglirans, and likely primitive for therians(Macrini et al., 2007), there are broad, round casts of the parafloc-cular lobes of the cerebellum present (Fig. 3A, B, D, F).

Brainstem and Cranial Nerves

The hypophyseal fossa can be observed on both virtual endo-casts (Fig. 3B, F), but is only well enough preserved to be mea-sured in USNM 530208, in which it is wider (width = 2.6 mm)than long (length = 1.7 mm), and reasonably deep (depth ∼1.8 mm). These proportions differ from those observed in bothM. annectens, in which the fossa is longer than wide and veryshallow (Silcox et al., 2010a), and in I. graybullianus, in which it isrelatively deep, but more nearly circular (Silcox et al., 2009). Al-though Macrini et al. (2007) scored characters related to the pro-portions of the hypophyseal fossa, this wide range of morpholo-gies in a group of fairly closely related taxa suggests that the sizeand shape of this fossa may be a feature of relatively low phyloge-netic valence. This inference should perhaps not be surprising be-cause the hypophyseal fossa’s form is apparently a poor predictorof the form of the pituitary gland in mammals (Edinger, 1942).

Although the optic chiasm is not preserved, casts of cranialnerve II are identifiable in the virtual endocast of USNM 530221rostral to the hypophyseal fossa (Fig. 3F). Lateral to the hypophy-seal fossa in the virtual endocast of USNM 530221 is an undiffer-entiated bulge that likely corresponds to the trigeminal ganglion,and the passageway of cranial nerves III, IV, V1, and V2 and theophthalmic vein to the sphenorbital fissure (Fig. 3F). Silcox et al.(2010b) suggested that there was no distinct foramen rotundum inLabidolemur kayi, so all of these vessels and nerves presumablypassed through this opening. This area is obstructed in ventralview in the virtual endocast of USNM 530208 by the cast of thealisphenoid canal (Fig. 3B), which is inferred to have housed theramus infraorbitalis of the ramus inferior of the stapedial artery(Silcox et al., 2010b). The cast of the foramen ovale, for cranial

nerve V3, is identifiable on the virtual endocast of USNM 530208,located just caudal to the alisphenoid canal (Figs. 3B, 5A). Bothvirtual endocasts exhibit clear casts of the internal auditory mea-tus, with distinct casts for nerves VII and VIII (Figs. 3B, F, 5B),just medial and ventral to the rostral end of the paraflocculus.In the virtual endocast of USNM 530208, the cast of the singlejugular (=posterior lacerate) foramen (which presumably housednerves IX, X, and XI, in addition to the sigmoid sinus) is evidentventral to the caudal end of the cast of the paraflocculus (Figs.3B, D, 5C). The cast of the single hypoglossal foramen for cra-nial nerve XII is only identifiable on the left side of the cast ofthe brainstem in the virtual endocast of USNM 530208 (Figs. 3B,5D).

Blood Vessels

Several elements of the system of sinuses and veins drain-ing blood from the brain are evident on the endocast of USNM530208, including the caudal portion of the superior sagittal sinus,transverse sinus, postglenoid vein, and sigmoid sinus (in the castof the jugular foramen; Figs. 3B, D, 5C). This suggests that L. kayiwould have had a fairly typical pattern of blood flow from thebrain, similar to that observed in I. graybullianus, M. annectens,and most other members of Euarchontoglires (except dermopter-ans, which lack the postglenoid vein). The superior sagittal si-nus is less pronounced rostrally (Fig. 3A, E), suggesting that itmay have been located deep within the meninges (Macrini et al.,2007); a similar morphology was observed in the endocasts of mi-crosyopids (Silcox et al., 2010a). Casts of the sigmoid and inferiorpetrosal sinuses are not clearly evident on either virtual endocast,but this is likely influenced by poor preservation in the relevantregions. There are no casts of foramina for rami temporales oremissary veins passing through the parietal, or along the parietal-squamosal suture in the virtual endocasts of L. kayi. As notedby Silcox et al. (2010b), the absence of these foramina contrastswith other apatemyids known from the cranium. Although Silcoxet al. (2010b) identified several landmarks of a transpromonto-rial internal carotid artery in the basicranium of USNM 530208,only a few parts of the path of arterial blood flow to the brainare evident on the virtual endocasts, including the orbitotemporalcanal (presumably for the ramus superior of the stapedial artery,in addition to one or more veins; Wible, 1987), and the alisphe-noid canal for the ramus infraorbitalis of the ramus inferior ofthe stapedial artery (Fig. 3B, D). No cast of the anterior carotidforamen for the promontory artery is evident on either of the vir-tual endocasts—this likely reflects the poor preservation of therelevant region in both specimens (Silcox et al., 2010b).

Brain Size and EQ

Obtaining an accurate estimate of cranial capacity is compli-cated by the fact that neither endocast for L. kayi is complete,and by the evident difference in size between the two individualsto which they pertained (i.e., the area of M1 in USNM 530221is 86% that of USNM 530208; based on data from Silcox et al.,2010b). Although it is generally preferable to minimize manip-ulation of the data for well-preserved specimens, in this case astraight overlaying of the two endocasts on one another usinga rigid warp algorithm produces a result that is clearly poorlyproportioned (Fig. 2B), and the volume estimate derived fromit is probably too large for the individual represented by USNM530221 and too small for the individual represented by USNM530208. The result of the Bookstein warp (Fig. 2C) appears to bea more reasonable representation of the size and proportions ofthe endocast of USNM 530221, but in the absence of any externalsource of validation this reconstruction must be treated with cau-tion. To be maximally conservative, we have included estimatesof volume based on four levels of manipulation of the data in Ta-ble 1. These different assumptions sets produce a range of cranial

Dow

nloa

ded

by [U

nive

rsity

of T

oron

to L

ibra

ries]

at 0

7:33

08

Nov

embe

r 201

1


capacity estimates that extends from approximately 0.5 to 0.6 cc.An estimate of the mass of the brain can be calculated by assum-ing that brain tissue has a density similar to liquid water (Gin-gerich and Gunnell, 2005), which has a specific gravity of 1 g/cm3.Using this assumption, the brain of L. kayi would have been ap-proximately 0.5–0.6 g (mass = volume/specific gravity).

Encephalization quotient (EQ) estimates calculated from allfour estimates, based on the body mass estimate for USNM530221 (74 g; Armstrong et al., 2011), are included in Table 1 andFigure 7. These EQ estimates lie within the range calculated forplesiadapiforms, and near the top part of the range of estimatesfor ‘archaic’ non-primates from the Late Cretaceous and earlyTertiary. However, they are markedly lower than EQ estimatesfor C. sigei calculated by Koenigswald et al. (2009; 0.81 using Jeri-son’s 1973 equation). Although this may stem in part from the dif-ferent bases for body mass estimation (teeth in C. sigei; postcra-

FIGURE 7. Box plot of encephalization quotients of living non-primateeuarchontoglirans, living and fossil non-hominin euprimates, archaic non-primate mammals, plesiadapiforms, Labidolemur kayi, and Carcinellasigei using Eisenberg’s (1981) equation. Note that the range of valuesfor Rodentia represents only two data points (Pirlot and Kamiya, 1982),and is therefore unlikely to be representative of the full range of varia-tion for the order. The archaic non-primate mammals include a variety oflate Cretaceous and early Tertiary taxa (see Supplementary Table 2). Therange of estimates portrayed for L. kayi represents varying estimates ofcranial capacity based on different assumption sets (see Table 1), whereasthe range provided for C. sigei reflects the prediction limits for the bodymass estimate calculated by Koenigswald et al. (2009). Data from Stephanet al., 1970; Szalay and Berzi, 1973; Radinsky, 1978; Pirlot and Kamiya,1982; Conroy, 1987 (just for Oreopithecus body mass estimate); Martin,1990; Simons and Rasmussen, 1996; Bush et al., 2004; Gingerich and Gun-nell, 2005; Sears et al., 2008; Koenigswald et al., 2009; Silcox et al., 2009,2010a; and the current study.

nials in L. kayi), a direct comparison of the data available fromboth specimens suggests that some portion of this difference isreal. The two species possessed generally similar dental morphol-ogy, which suggests that the teeth likely scaled in a similar man-ner in relation to body mass. The area of M1 is smaller in C. sigei(2.77 mm2; based on data from Koenigswald et al., 2009) than foreither specimen of L. kayi (USNM 53028 = 4.23 mm2; USNM530221 = 3.65 mm2 based on data from Silcox et al., 2010b), butthe calculated cranial capacity of C. sigei (1.3 cc) is more thantwice the maximum estimate produced for L. kayi (Table 1). Thisstrongly suggests that there was a significant difference in relativebrain size between the two species, independent of the methodused for body mass estimation.

DISCUSSION

The most notable contrast between the endocast of Labidole-mur kayi, and those of non-apatemyid, early Tertiary euar-chontoglirans, is in the size of the olfactory bulbs (Fig. 4).Jerison (1973) argued that large olfactory bulbs represent theprimitive condition for mammals, and this inference is sup-ported by the relatively large size of this region of the brainin Cretaceous eutherians (Kielan-Jaworowska, 1984; Supple-mentary Table 1; supplementary materials available onlineat www.vertpaleo.org/jvp/JVPcontents.html), and by Macriniet al.’s (2007) character analysis. Macrini et al. (2007) used a cut-off of 6% for percentage of the endocast composed of the olfac-tory bulb casts to differentiate between ‘large’ and ‘small.’ Basedon this cutoff, all living and fossil euarchontans known from en-docasts, including plesiadapiforms, would fall into the ‘small’ sizerange (Supplementary Table 1; based on data from Stephan et al.,1970, 1981; Gurche, 1982; Martin, 1990). There are much fewerpublished data for members of Glires, but Pirlot and Kamiya(1982) provide volumes for two genera of gliding rodents (Iomysand Glaucomys), who both also fall well below 6% (Supplemen-tary Table 1), and both living and fossil erethizontid rodents havetiny olfactory bulbs (Dozo et al., 2004). In contrast, the two ap-atemyids, Carcinella sigei and L. kayi, appear to share relativelylarge olfactory bulbs, with all estimates for L. kayi substantiallyexceeding the 6% cutoff (Table 1). This suggests that apatemyidsmay have been primitive relative to other euarchontoglirans inretaining relatively large olfactory bulbs. Assuming the hypoth-esis of relationships provided in Figure 1 is correct, it is possiblethat reduction of this region occurred independently in euarchon-tans and glirans.

One caveat to this interpretation is the recognition that sometaxa may evolve relatively larger olfactory bulbs in associationwith specializations to the sense of smell—for example, Novacek(1982) suggested this for certain eulipotyphlans. It has been sug-gested that apatemyids were similar in their ecology to extant ex-tractive foragers such as Daubentonia madagascarensis and thefour species of petaurid marsupial Dactylopsila (Koenigswaldand Schierning, 1987). Living aye-ayes have an exceptionallycomplex nasal skeleton compared to other strepsirrhines (Maier,1993; Ruf, 2008), and even though their olfactory bulbs accountfor only approximately 1.6% of their total endocranial volume(Stephan et al., 1970), these structures are relatively enlargedcompared to those of other euprimates, whereas areas of thebrain associated with vision are reduced (Kaufman et al., 2005).Therefore, it is possible that the large size of the olfactory bulbs inapatemyids reflect, at least in part, similar specializations relatedto extractive foraging.

The similarity in the inferred position of the rhinal sulcus in Ig-nacius graybullianus, Microsyops annectens, and L. kayi suggeststhat this state may be primitive for Euarchontoglires, although inthe absence of other additional relevant data (e.g., from primi-tive scandentians, dermopterans, or members of Glires) this con-clusion remains tentative. This position is more ventral than that

Dow

nloa

ded

by [U

nive

rsity

of T

oron

to L

ibra

ries]

at 0

7:33

08

Nov

embe

r 201

1


observed in Cretaceous eutherians (Kielan-Jaworowska, 1984;but see Macrini et al., 2007) and living insectivorans that havebeen considered to have relatively primitive cerebral patternssuch as Solenodon (see Allen, 1910: pl. 6; e.g., ‘basal insectivores’of Stephan et al., 1970; Stephan, 1972), suggesting some neocorti-calization may have occurred in the early evolution of Euarchon-toglires.

The virtual endocasts of L. kayi share a number of characteris-tics with those published for plesiadapiforms (Fig. 3). In additionto the approximate position of the orbitotemporal canal, theseinclude exposed caudal colliculi (missing in some derived mem-bers of Microsyopidae, but inferred to be primitive for the family;Silcox et al., 2010a; the midbrain is also exposed in Rhombomy-lus turpanensis—see Meng et al., 2003:fig. 50) and an overlappingrange of EQ estimates. Although it would be beneficial to assessthese features in a wider range of early Tertiary euarchontogli-rans, if they were available, the distribution of these similaritiesnonetheless suggests that they may have characterized the com-mon ancestor of Euarchontoglires.

Features observed in fossil euprimate endocasts that are miss-ing in both L. kayi and in plesiadapiforms include the sylviansulcus (missing in Smilodectes; Gazin, 1965), caudal expansion ofthe cerebrum to cover the midbrain, and a well-developed tem-poral pole. The absence of these features in an additional groupwith possible links to Euarchonta supports the inference that theycould be euprimate synapomorphies (Silcox et al., 2010a). Eventhough more has been written about the evolution of the brainin Primates than in any other order, one of the problems thatremains in answering questions about early primate brain evolu-tion is the absence of closely related fossil outgroups to Primatessensu lato (i.e., including plesiadapiforms), to establish what isprimitive for the group as a whole. Although extant treeshrewshave often been used as stand-ins for primitive primates (e.g., LeGros Clark, 1924, 1959; Kaas 2008), contrasts between the formof the brain in tupaiid scandentians and the endocasts of plesi-adapiforms suggest that the brain of Tupaia, at least, makes apoor model for the primitive morphology of the brain in Eupri-mates (Silcox et al., 2009, 2010a). This viewpoint is supported byanalyses that suggest that specializations of the brain associatedwith diurnal visual processing may have evolved independentlyin the two orders (Campbell, 1966, 1980; Allman, 1977), with Tu-paia actually having a visual system more similar to that of livingsciurid rodents than to primates in some features (Kaas, 2002;but see Lyon, 2009). Contrasts in the form of the brain betweenTupaia and Ptilocercus, including, for example, the relativelysmaller rostral (superior) colliculi in the latter (Le Gros Clark,1926), also support the inference that many of the specializa-tions of the visual system of diurnal treeshrews occurred withinScandentia. Another frequently adopted approach uses eulipo-typhlans such as Solenodon as stand-ins for primitive primates(e.g., Stephan et al., 1970, 1982; Martin, 1990:fig. 8.17). An obvi-ous problem with this approach is the great phylogenetic distancebetween eulipotyphlans and primates in current, broadly heldhypotheses of mammalian interrelationships (e.g., see Springeret al., 2004).

Extinct apatemyids, in particular L. kayi, may form a more ap-propriate basis for comparison with primates than these extanttaxa. For example, the similarity in the position of the orbitotem-poral canal in L. kayi and plesiadapiforms suggests that this po-sition might be primitive for Euarchontoglires. If this structure isassociated with the rhinal fissure in these forms, then this wouldrepresent a more ventral position for this critical dividing line be-tween the neo- and paleo-cortices than observed in primitive eu-lipotyphlans, implying a more ancient history for at least somemeasure of neocorticalization. One thing worthy of note, how-ever, is that the relatively larger olfactory bulbs in a species witha similar EQ to those calculated for plesiadapiforms implies thatthe rest of the brain was relatively smaller in L. kayi than in prim-

itive primates, suggesting perhaps a relatively larger neocortexwas present in plesiadapiforms.

In terms of the very small size of the olfactory bulbs observedin modern euprimates, the contrast between the large bulbsof apatemyids, and the smaller bulbs of other non-euprimateeuarchontoglirans, suggests that the acquisition of Euprimates’relatively diminutive olfactory bulbs may have occurred in amulti-step process, with some reduction in the size of this re-gion (or expansion to other parts of this brain) pre-dating theeuprimate node. This further emphasizes the inappropriatenessof using living eulipotyphlans, with their larger olfactory bulbs,as primitive models for the brain in early primates.

The endocasts of L. kayi are also relevant for considering evo-lution of the brain within Apatemyidae. The relative brain sizeof C. sigei is large compared not only to L. kayi, but to other ‘ar-chaic’ mammals (Fig. 7). Some authors (e.g., Gibson, 1986) havesuggested that complex extractive feeding behaviors may haveplayed a role in the evolution of primate intelligence, and indeedDaubentonia is the most encephalized of the lemurs (Barrick-man and Lin, 2010), with an enlarged frontal cortex (Kaufmanet al., 2005). Perhaps the increase in relative brain size that seemsto have occurred in Apatemyidae might also be tied to elab-orations in foraging behavior. Koenigswald et al. (2005a) doc-umented more specialized hand proportions in middle EoceneHeterohyus nanus than observed in early Eocene Apatemys char-dini or L. kayi, which is consistent with the idea that apatemyidsbecame more specialized for manual extractive foraging throughtime. This hypothesis would require the discovery of postcranialmaterial for C. sigei to be tested.

CONCLUSIONS

The endocranial anatomy of Labidolemur kayi is interestingboth from the perspective of its potential importance to under-standing euarchontogliran brain evolution, and to informing ourunderstanding of evolution within the Apatemyidae. Featuresshared with other early Tertiary euarchontoglirans that may beprimitive for the group include exposed caudal colliculi, a simi-larly located orbitotemporal canal (that may reflect the locationof the rhinal sulcus) on the lateral surface of the cerebrum, andan overlapping range of EQ estimates. The large size of the olfac-tory bulbs contrasts with the condition in other early Tertiary eu-archontoglirans, excepting the only other apatemyid known froman endocast, Carcinella sigei. However, the endocast of L. kayicontrasts markedly with that of C. sigei in being much smaller,both absolutely, and relative to body mass. This suggests signif-icant increase occurred in brain size in Apatemyidae over thecourse of the Eocene, which might be related to elaborations inthe family’s specialized manual extractive feeding regime.

ACKNOWLEDGMENTS

We thank A. Walker, P. Halleck, A. Grader, O. Karacan, andT. Ryan (Penn State, Center for Quantitative Imaging) for helpwith CT scanning and P. D. Gingerich (University of Michigan)for access to material. This research was supported by NSF re-search grants BCS-0003920 to A. Walker and F. Spoor and DEB-0629836 to M.T.S., J.I.B., and E. Sargis, and an NSERC discoverygrant to M.T.S. Thanks to reviewers I. Ruf and J. Wible for com-ments that substantively improved the manuscript.

LITERATURE CITED

Allen, G. M. 1910. Solenodon paradoxus. Memoirs of the Museum ofComparative Zoology at Harvard College 40:1–54, pl. 1–9.

Allman, J. M. 1977. Evolution of the visual system in the early primates;pp. 1–53 in J. M. Sprague and A. N. Epstein (eds.), Progress in Psy-chobiology: Physiology and Psychology, Volume 7. Academic Press,New York, New York.

Dow

nloa

ded

by [U

nive

rsity

of T

oron

to L

ibra

ries]

at 0

7:33

08

Nov

embe

r 201

1


Armstrong, S. D., J. I. Bloch, P. Houde, and M. T. Silcox. 2011. Cochlearlabyrinth volume in euarchontoglirans: implications for the evolu-tion of hearing in Primates. Anatomical Record 294:263–266.

Barrickman, N. L., and M. J. Lin. 2010. Encephalization, expensive tis-sues, and energetics: an examination of the relative costs of brainsize in strepsirrhines. American Journal of Physical Anthropology143:579–590.

Bloch, J. I., D. M. Boyer, M. T. Silcox, and P. Houde. 2004. New skeletonsof Paleocene-Eocene Labidolemur kayi (Mammalia, Apatemyidae):ecomorphology and relationship of apatemyids to Primates andother mammals. Journal of Vertebrate Paleontology 24(3, Supple-ment):40A.

Bloch, J. I., M. T. Silcox, D. M. Boyer, and E. J. Sargis. 2007. New Pa-leocene skeletons and the relationship of plesiadapiforms to crown-clade primates. Proceedings of the National Academy of Science ofthe United States of America 104:1159–1164.

Bush, E. C., E. L. Simons, D. J. Dubowitz, and J. M. Allman. 2004. En-docranial volume and optic foramen size in Parapithecus grangeri;pp. 603–614 in C. F. Ross and R. F. Kay (eds.), Anthropoid Origins:New Visions. Kluwer, Boston, Massachusetts.

Campbell, C. B. G. 1966. The relationships of the tree shrews: the evi-dence of the nervous system. Evolution 20:276–281.

Campbell, C. B. G. 1980. The nervous system of the Tupaiidae: its bear-ing on phyletic relationships; pp. 219–242 in W. P. Luckett and C.R. Noback (eds.), Comparative Biology and Evolutionary Relation-ships of Tree Shrews. Plenum, New York, New York.

Conroy, G. C. 1987. Problems of body-weight estimation in fossil pri-mates. International Journal of Primatology 8:115–137.

Dozo, M. T., M. G. Vucetich, and A. M. Candela. 2004. Skull anatomyand neuromorphology of Hypsosteiromys, a Colhuehuapian erethi-zontid rodent from Argentina. Journal of Vertebrate Paleontology24:228–234.

Edinger, T. 1942. The pituitary body in giant animals fossil and living: asurvey and a suggestion. Quarterly Review of Biology 17:31–54.

Edinger, T. 1964. Midbrain exposure and overlap in mammals. AmericanZoologist 4:5–19.

Eisenberg, J. F. 1981. The Mammalian Radiations: An Analysis of Trendsin Evolution, Adaptation, and Behavior. University of ChicagoPress, Chicago, Illinois, 610 pp.

Gazin, C. L. 1965. An endocranial cast of the Bridger middle Eoceneprimate Smilodectes gracilis. Smithsonian Miscellaneous Collections149:1–14.

Gibson, K. R. 1986. Cognition, brain size and the extraction of embeddedfood resources; pp. 93–104 in J. Else and P. C. Lee (eds.), PrimateOntogeny, Cognition and Social Behaviour. Cambridge UniversityPress, Cambridge, U.K.

Gingerich, P. D. 1982. Studies on Paleocene and early Eocene Apatemyi-dae (Mammalia, Insectivora). II. Labidolemur and Apatemys fromthe early Wasatchian of the Clark’s Fork Basin, Wyoming. Contri-butions from the Museum of Paleontology, University of Michigan26:57–69.

Gingerich, P. D. 2003. Mammalian responses to climate change at thePaleocene-Eocene boundary: Polecat Bench record in the north-ern Bighorn Basin, Wyoming; pp. 463–478 in S. L. Wing, P. D Gin-gerich, B. Schmitz, and E. Thomas (eds.), Causes and Consequencesof Globally Warm Climates in the Early Paleogene. Geological So-ciety of America, Special Papers 369.

Gingerich, P. D., and G. F. Gunnell, 2005. Brain of Plesiadapis cookei(Mammalia, Proprimates): surface morphology and encephalizationcompared to those of Primates and Dermoptera. Contributions fromthe Museum of Paleontology, University of Michigan 31:185–195.

Gingerich, P. D., and R. D. Martin. 1981. Cranial morphology and adap-tations in Eocene Adapidae. II. The Cambridge skull of Adapisparisiensis. American Journal of Physical Anthropology 56:235–257.

Gurche, J. A. 1982. Early primate brain evolution; pp. 227–246 in E. Arm-strong and D. Falk (eds.), Primate Brain Evolution: Methods andConcepts. Plenum, New York, New York.

Heller, F. 1930. Die Saugetierfauna der mitteleozanen Braunkohle desGeiseltales bei Halle a. S. Jahrbuch des Halleschen Verbandes furdie Erforschung der mitteldeutschen Bodenschatze und ihrer Verw-ertung 9:13–41.

Jepsen, G. L. 1934. A revision of the American Apatemyidae and thedescription of a new genus, Sinclairella, from the White RiverOligocene of South Dakota. Proceedings of the American Philo-sophical Society 74:287–305.

Jerison, H. J. 1973. Evolution of the Brain and Intelligence. AcademicPress, New York, New York, 482 pp.

Jerison, H. J. 1979. Brain, body and encephalization in early primates.Journal of Human Evolution 8:615–635.

Jerison, H. J. 1991. Fossil brains and the evolution of the neocortex; pp.5–19 in B. L. Finlay, G. Innoncenti, and G. Scheich (eds.), The Neo-cortex: Ontogeny and Phylogeny. Nato Advanced Science InstitutesSeries A: Life Sciences, Volume 200. Plenum Press, New York, NewYork.

Kaas, J. H. 2002. Convergences in the modular and areal organizationof the forebrain of mammals: implications for the reconstruction offorebrain evolution. Brain Behavior and Evolution 59:262–272.

Kaas, J. H. 2008. The evolution of the complex sensory and motor systemsof the human brain. Brain Research Bulletin 75:384–390.

Kalthoff, D. C., W. von Koenigswald, and C. Kurz. 2004. A new specimenof Heterohyus nanus (Apatemyidae, Mammalia) from the Eoceneof Messel (Germany) with unusual soft part preservation. CourierForschungsinstitut Senckenberg 252:1–12.

Kaufman, J. A., E. T. Ahrens, D. H. Laidlaw, S. Zhang, and J. M. All-man. 2005. Anatomical analysis of an aye-aye brain (Daubentoniamadagascariensis, Primates: Prosimii) combining histology, struc-tural magnetic resonance imaging, and diffusion-tensor imaging.Anatomical Record Part A 287:1026–1037.

Kielan-Jaworowska, Z. 1984. Evolution of the therian mammals in theLate Cretaceous of Asia. Part VI. Endocranial casts of eutherianmammals. Acta Palaeontologica Polonica 46:157–171.

Koenigswald, W. von. 1987. Apatemyiden-Skelette aus dem Mitteleozanvon Messel und ihre palaobiologische Aussage. Carolinea 45:31–35.

Koenigswald, W. von. 1990. Die Palaobiologie der Apatemyiden (Insec-tivora s.l.) und die Ausdeutung der Skelettfunde von Heterohyusnanus aus dem Mitteleozan von Messel bei Darmstadt. Palaeonto-graphica A 210:41–77.

Koenigswald, W. von, and H.-P. Schierning. 1987. The ecological nicheof early Tertiary apatemyids—extinct group of mammals. Nature326:595–596.

Koenigswald, W. von, I. Ruf, and P. D Gingerich. 2009. Cranial morphol-ogy of a new apatemyid, Carcinella sigei n. gen. n. sp. (Mammalia,Apatotheria) from the late Eocene of southern France. Paleonto-graphica A 288:53–91.

Koenigswald, W. von, K. D. Rose, L. Grande, and R. D. Martin. 2005a.First apatemyid skeleton from the lower Eocene Fossil Butte Mem-ber, Wyoming (USA), compared to the European apatemyid fromMessel, Germany. Palaeontographica A 272:149–169.

Koenigswald, W. von, K. D. Rose, L. Grande, and R. D. Martin. 2005b.Die Lebensweise eozaner Saugetiere (Pantolestidae und Apatemyi-dae) aus Messel (Europa) im Vergleich zu neuen Skelettfunden ausdem Fossil Butte Member von Wyoming (Nordamerika). Geologis-ches Jahrbuch Hessen 272:43–54.

Le Gros Clark, W. E. 1924. On the brain of the tree-shrew (Tu-paia minor). Proceedings of the Zoological Society of London 94:559–567.

Le Gros Clark, W. E. 1926. On the anatomy of the pen-tailed tree-shrew(Ptilocercus lowii). Proceedings of the Zoological Society of London96:1179–1309.

Le Gros Clark, W. E. 1932. The brain of the Insectivora. Proceedings ofthe Zoological Society of London 1932:975–1013.

Le Gros Clark, W. E. 1945. Note on the palaeontology of the lemuroidbrain. Journal of Anatomy 79:123–126.

Le Gros Clark, W. E. 1959. The Antecedents of Man. Edinburgh Univer-sity Press, Edinburgh, U.K., 374 pp.

Lyon, D. C. 2009. The evolution of visual cortex and visual systems; pp.751–790 in J. H. Kaas (ed.), Evolutionary Neuroscience. AcademicPress, San Diego, California.

MacPhee, R. D. E., M. Cartmill, P. D. Gingerich. 1983. New Paleogeneprimate basicrania and the definition of the order Primates. Nature301:509–511.

Macrini, T. E., G. W. Rougier, and T. Rowe. 2007. Description of a cranialendocast from the fossil mammal Vincelestes neuquenianus (Theri-iformes) and its relevance to the evolution of endocranial charactersin therians. Anatomical Record 290:875–892.

Maier, W. 1993. Zur evolutiven und funktionellen Morphologie desGesichtsschadels der Primaten. Zeitschrift fur Morphologie undAnthropologie 79:279–299.

Martin, R. D. 1990. Primate Origins and Evolution: A Phylogenetic Re-construction. Chapman and Hall, London, 804 pp.

Dow

nloa

ded

by [U

nive

rsity

of T

oron

to L

ibra

ries]

at 0

7:33

08

Nov

embe

r 201

1


Matthew, W. D. 1909. The Carnivora and Insectivora of the BridgerBasin, middle Eocene. Memoirs of the American Museum of Natu-ral History 9:291–567.

Matthew, W. D. 1921. Stehlinius, a new Eocene insectivore. AmericanMuseum Novitates 14:1–5.

Matthew, W. D. 1929. Preoccupied names. Journal of Mammalogy 10:171.McKenna, M. C. 1963. Primitive Paleocene and Eocene Apatemyi-

dae (Mammalia, Insectivora) and the primate-insectivore boundary.American Museum Novitates 2160:1–39.

McKenna, M. C., and S. K. Bell. 1997. Classification of Mammals abovethe Species Level. Columbia University Press, New York, NewYork, 631 pp.

Meng, J., Y. Hu, and C-K. Li. 2003. The osteology of Rhombomy-lus (Mammalia, Glires): implications for phylogeny and evolutionof Glires. Bulletin of the American Museum of Natural History275:1–247.

Moodie, R. L. 1922. On the endocranial anatomy of some Oligoceneand Pleistocene mammals. Journal of Comparative Neurology34:348–371.

Novacek, M. J. 1982. The brain of Leptictis dakotensis, an Oligocene lep-tictid, Eutheria, Mammalia, from North America. Journal of Pale-ontology 56:1177–1186.

Pirlot, P., and T. Kamiya. 1982. Relative size of brain and brain compo-nents in three gliding placentals (Dermoptera; Rodentia). CanadianJournal of Zoology 60:565–572.

Preuss, T. M. 2009. Primate brain evolution; pp. 794–825 in J. H. Kaas(ed.), Evolutionary Neuroscience. Academic Press, San Diego,California.

Radinsky, L. B. 1967. The oldest primate endocast. American Journal ofPhysical Anthropology 27:385–388.

Radinsky, L. B. 1970. The fossil evidence of prosimian brain evolution;pp 209–224 in C. R. Noback and W. Montagna (eds.), The Pri-mate Brain: Advances in Primatology, Volume 1. Appleton CenturyCrofts, New York, New York.

Radinsky, L. B. 1975. Primate brain evolution. American Scientist63:656–663.

Radinsky, L. B. 1977. Early primate brains: facts and fiction. Journal ofHuman Evolution 6:79–86.

Radinsky, L. B. 1978. Evolution of brain size in carnivores and ungulates.American Naturalist 112:815–831.

Rasband, W. S. 1997–2008. ImageJ. U.S. National Institutes of Health,Bethesda, Maryland. Available at http://rsb.info.nih.gov/ij/. Ac-cessed June 21, 2010.

Romer, A. S. 1945. Vertebrate Paleontology. University of Chicago Press,Chicago, Illinois, 687 pp.

Ruf, I., P. D. Gingerich, and W. von Koenigswald. 2008. Do internal cra-nial features reflect adaptations in “woodpecking” mammals. Jour-nal of Vertebrate Paleontology 28(3, Supplement):134A.

Schlosser, M. 1918. Klasse Mammalia; pp. 380–659 in K. A. von Zittel(ed.), Grundzuge der Palaontologie. II Abteilung: Vertebrata. 3rdedition. R. Oldenbourg, Munich and Berlin, Germany.

Scott, W. B., and G. L. Jepsen. 1936. The mammalian fauna of the WhiteRiver Oligocene. Part I. Insectivora and Carnivora. Transactions ofthe American Philosophical Society New Series 28:1–153.

Sears, K. E., J. A. Finarelli, J. J. Flynn, and A. R. Wyss. 2008. Estimatingbody mass in New World “monkeys” (Platyrrhini, Primates), with aconsideration of the Miocene platyrrhine, Chilecebus carrascoensis.American Museum Novitates 3617:1–29.

Silcox, M. T., A. E. Benham, and J. I. Bloch. 2010a. Endocasts of Mi-crosyops (Microsyopidae, Primates) and the evolution of the brainin primitive primates. Journal of Human Evolution 58:505–521.

Silcox, M. T., C. K. Dalmyn, and J. I. Bloch. 2009. Virtual endocast of Ig-nacius graybullianus (Paromomyidae, Primates) and brain evolutionin early Primates. Proceedings of the National Academy of Sciencesof the United States of America 106:10987–10992.

Silcox, M. T., J. I. Bloch, D. M. Boyer, and P. Houde. 2010b. Cra-nial anatomy of Paleocene and Eocene Labidolemur kayi (Mam-malia: Apatotheria) and the relationships of the Apatemyidae toother mammals. Zoological Journal of the Linnean Society 160:773–825.

Simons, E. L., and D. T. Rasmussen. 1996. The skull of Catopithecusbrowni, an early Tertiary catarrhine. American Journal of PhysicalAnthropology 100:261–292.

Simpson, G. G. 1940. Studies on the earliest primates. Bulletin of theAmerican Museum of Natural History 77:185–212.

Simpson, G. G. 1945. The principles of classification and a classificationof mammals. Bulletin of the American Museum of Natural History85:1–350.

Springer, M. S., M. J. Stanhope, O. Madsen, and W. W. deJong. 2004.Molecules consolidate the placental mammal tree. Trends in Ecol-ogy and Evolution 19:430–437.

Stehlin, H. G. 1916. Die Saugetiere des Schweizerischen Eocaens. Ab-handlungen der Schweizerischen Palaontologischen Gesellschaft41:1299–1552.

Stephan, H. 1972. Evolution of primate brains: a comparative anatom-ical investigation; pp. 155–174 in R. Tuttle (ed.), The Functionaland Evolutionary Biology of Primates. Aldine-Atherton, New York,New York.

Stephan, H., R. Bauchot, and O. J. Andy. 1970. Data on size of the brainand of various brain parts in insectivores and primates; pp 289–297in C. R. Noback and W. Montagna (eds.), The Primate Brain: Ad-vances in Primatology, Volume 1. Appleton Century Crofts, NewYork, New York.

Stephan, H., H. Frahm, and G. Baron. 1981. New and revised data onvolumes of brain structures in insectivores and primates. Folia Pri-matologica 35:1–29.

Szalay, F. S. 1968. Origins of the Apatemyidae (Mammalia, Insectivora).American Museum Novitates 2352:1–11.

Szalay, F. S. 1969. Mixodectidae, Microsyopidae, and the insectivore-primate transition. Bulletin of the American Museum of NaturalHistory 140:195–330.

Szalay, F. S., and A. Berzi. 1973. Cranial Anatomy of Oreopithecus. Sci-ence 180:183–185.

Teilhard de Chardin, P. 1922. Les mammiferes de l’Eocene inferieurfrancais et leurs gisements. Annales de Paleontologie, Paris11:9–116.

West, R. M. 1973. Review of the North American Eocene and OligoceneApatemyidae (Mammalia: Insectivora). Special Publications theMuseum Texas Tech University 3:1–42.

West, R. M., and E. G. Atkins. 1970. Additional middle Eocene (Bridge-rian) mammals from Tabernacle Butte, Sublette County, Wyoming.American Museum Novitates 2404:1–26.

Wible, J. R. 1987. The eutherian stapedial artery: character analysis andimplications for superordinal relationships. Zoological Journal ofthe Linnean Society 91:107–135.

Wible, J. R. 2008 On the cranial osteology of the Hispaniolansolenodon, Solenodon paradoxus Brandt, 1833 (Mammalia, Lipo-typhla, Solenodontidae). Annals of Carnegie Museum 77:321–402.

Wible, J. R., and T. J. Gaudin. 2004. On the cranial osteology of the yellowarmadillo Euphractus sexcinctus (Dasypodidae, Xenarthra, Placen-talia). Annals of Carnegie Museum 73:117–196.

Wood, A. E. 1937. The mammalian fauna of the White River Oligocene.Part II. Rodentia. Transactions of the American Philosophical Soci-ety 28:157–269.

Submitted December 1, 2010; revisions received June 27, 2011; acceptedJuly 28, 2011.Handling editor: Thomas Martin.

Dow

nloa

ded

by [U

nive

rsity

of T

oron

to L

ibra

ries]

at 0

7:33

08

Nov

embe

r 201

1

Endocranial morphology of Labidolemur kayi (Apatemyidae

Documents

Transcript of Endocranial morphology of Labidolemur kayi (Apatemyidae