Citation
The First Virtual Endocasts of North American Adapiform Primates

Material Information

Title:
The First Virtual Endocasts of North American Adapiform Primates
Creator:
Harrington, Arianna R
Place of Publication:
[Gainesville, Fla.]
Florida
Publisher:
University of Florida
Publication Date:
Language:
english
Physical Description:
1 online resource (123 p.)

Thesis/Dissertation Information

Degree:
Master's ( M.S.)
Degree Grantor:
University of Florida
Degree Disciplines:
Zoology
Biology
Committee Chair:
BLOCH,JONATHAN I
Committee Co-Chair:
MACFADDEN,BRUCE J
Committee Members:
DAEGLING,DAVID
Graduation Date:
5/2/2015

Subjects

Subjects / Keywords:
Anthropometric measurements ( jstor )
Average linear density ( jstor )
Cerebrum ( jstor )
Diameters ( jstor )
Estimated taxes ( jstor )
Fossils ( jstor )
Mammals ( jstor )
Olfactory bulb ( jstor )
Primates ( jstor )
Specimens ( jstor )
Biology -- Dissertations, Academic -- UF
adapiformes -- anatomy -- brain -- endocast -- evolution -- notharctidae -- notharctus -- primate -- smilodectes
Genre:
bibliography ( marcgt )
theses ( marcgt )
government publication (state, provincial, terriorial, dependent) ( marcgt )
born-digital ( sobekcm )
Electronic Thesis or Dissertation
Zoology thesis, M.S.

Notes

Abstract:
Notharctidae is an Eocene North American adapiform euprimate (crown primate) with a plentiful fossil record, making examination of their endocasts ideal for inferring the neuroanatomy and encephalization of stem strepsirrhines. Virtual endocasts of the notharctids Notharctus tenebrosus (n=3) and Smilodectes gracilis (n=3) from the Bridger formation of Wyoming and the late Eocene European adapid adapiform Adapis parisiensis (n=1) were reconstructed from high-resolution X-ray computed tomography (CT) scan data. While the three species shared many neuroanatomical similarities differentiating them from plesiadapiforms (stem primates) and extant euprimates, cerebral proportions of N. tenebrosus were more variable than S. gracilis, which may be related to each specimen's stratigraphic position and each species' level of cranial sexual dimorphism. One specimen of S. gracilis preserves the first record of a ?Sylvian sulcus for this species. Dental body mass estimates suggest that the encephalization quotient (EQ) of N. tenebrosus was higher than S. gracilis, which is surprising as they are inferred to have shared common diets, activity patterns, and inferred social structures, and are purportedly closely related. However, body mass estimates using postcranial variables suggest that some specimens of both species were of similar body size and EQ, suggesting that dental proxies overestimate the body mass of N. tenebrosus, which was likely about as encephalized as or slightly less encephalized than S. gracilis. Regardless, the results of this study suggest that the EQ of notharctids were not much higher than those of plesiadapiforms and most likely did not overlap those of extant strepsirrhines and haplorhines. ( en )
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis:
Thesis (M.S.)--University of Florida, 2015.
Local:
Adviser: BLOCH,JONATHAN I.
Local:
Co-adviser: MACFADDEN,BRUCE J.
Electronic Access:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2016-05-31
Statement of Responsibility:
by Arianna R Harrington.

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Embargo Date:
5/31/2016
Classification:
LD1780 2015 ( lcc )

Downloads

This item has the following downloads:


Full Text

PAGE 1

THE FIRST VIRTUAL ENDOCASTS OF NORTH AMERICAN ADAPIFORM PRIMATES By ARIANNA R. HARRINGTON A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF THE MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2015

PAGE 2

© 2015 Arianna R. Harrington

PAGE 3

3 ACKNOWLEDGEMENTS I would like to full heartedly thank Dr. Jonathan Bloch of the Florida Museum of Natural History for serving as my primary advisor and committee chair, and Dr. Bruce MacFadden of the Florida Museum of Natural History and Dr. David Daegling of the Department of Anthropology at the University of Florida for serving on my committee. I would like to thank all three of my committee member s for their time and helpful suggestions with my project and this manuscript. I would also like to thank Dr. Mary Silcox of the University of Toronto, Scarborough, for her help, advice, and training with regards to the reconstruction of virtual endocasts, Dr. Douglas Boyer of Duke University for help acquiring, scanning, and facilitating access to specimens used for this study, and Gabriel Yapuncich of Duke University for helpful conversations about body mass estimation. I would also like to thank my fam ily and friends for their support during the course of my education.

PAGE 4

4 TABLE OF CONTENTS page ACKNOWLEDGEMENTS ................................ ................................ ................................ ............. 3 LIST OF TABLES ................................ ................................ ................................ ........................... 6 LIST OF FIGURES ................................ ................................ ................................ ......................... 7 ABSTRACT ................................ ................................ ................................ ................................ ..... 9 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .................. 11 Taxonomic and Phylogenetic Overview ................................ ................................ ................. 11 Endocasts ................................ ................................ ................................ ................................ 13 Institutional Abbreviations ................................ ................................ ................................ ..... 16 2 MATERIALS AND METHODS ................................ ................................ ........................... 18 3 RESULTS ................................ ................................ ................................ ............................... 27 Developmental Stage of UM 32773 ................................ ................................ ....................... 27 Sex of USNM V 23277, USNM V 23278, and USNM V 17996 ................................ ........... 28 Description of the Endocasts of Notharctus tenebrosus ................................ ......................... 29 Comparisons to Previously Published Endocasts of Notharct us tenebrosus .......................... 34 Description of the Virtual Endocasts of Smilodectes gracilis ................................ ................ 37 Comparisons to Previously Published Endocasts of Smilodectes gracilis ............................. 40 The Virtual Endocast of Adapis parisiensis ................................ ................................ ........... 43 Body Mass Estimates and Encephalization Quotients ................................ ............................ 43 Relative Sizes of the Olfactory Bulbs ................................ ................................ ..................... 46 4 DISCUSSION ................................ ................................ ................................ ......................... 76 Sylvian Sulcus ................................ ................................ ................................ ........................ 77 Suprasylvian Sulcus ................................ ................................ ................................ ................ 80 Cerebral Proportions and Sexual Dimorphism ................................ ................................ ....... 80 Ontogeny and Endocast Morphology ................................ ................................ ..................... 82 Stratigraphy and Endocast Morphology ................................ ................................ ................. 84 Is S. gracilis More Encephalized than N. tenebrosus ? ................................ ........................... 85 5 CONCLUSIONS ................................ ................................ ................................ .................... 91 APPENDIX A TABLES ................................ ................................ ................................ ................................ . 94

PAGE 5

5 B CODE ................................ ................................ ................................ ................................ ... 105 LIST OF REFERENCES ................................ ................................ ................................ ............. 115 BIOGRAPHICA L SKETCH ................................ ................................ ................................ ....... 123

PAGE 6

6 LIST OF TABLES Table page 2 1 High resolution CT scanner settings for each specimen in the sample. ............................. 26 2 2 Summary of equations used in calculations in this study ................................ .................. 26 3 1 Dental measurements of specimens in the sample. ................................ ............................ 47 3 2 Measurements of the virtual endocasts of Notharctus tenebrosus. ................................ .... 53 3 3 Measurements of the virtual endocasts of S milodectes gracilis ................................ ........ 58 3 4 Measu rements of the virtual endocast of Adapis parisiensis. ................................ ............ 62 3 5 Body mass estimates for specimens in the sample using dental proxies. .......................... 63 3 6 Measurements of postcranial variables of specimens in the sample. ................................ . 65 3 7 Body mass estimates of specimens in the sample based on equations from Gin gerich (1990). ................................ ................................ ................................ ................................ 65 3 8 Body mass estimates of specimens in the sample bas ed on equations from Boyer (2009 ). ................................ ................................ ................................ ................................ 66 3 9 Endocast volume and encephalization quotients for specimens in the sample. ................. 71 A 1 Canine dimensions of Notharctus tenebrosus and Smilodectes gracilis ........................... 94 A 2 Previous body mass, endocranial volume, encephalization quotient, and calculation method for fossil primates. ................................ ................................ ................................ 95 A 3 Average body mass, endocranial volume, encephalization quotient o f extant primate species. ................................ ................................ ................................ ............................... 96 A 4 Body mass, endocranial volume, and olfactory bulb volume of primates and ................................ ................................ ................................ .................. 103

PAGE 7

7 LIST OF FIGURES Figure page 1 1 A graphic representation of the time periods in which the genera Notharctus and Smilodectes and the species Notharctus tenebrosus and Smilodectes gracilis were present in North Americ a.. ................................ ................................ ................................ . 17 2 1 Occlusal view of the cranium of UM32773 and a CT cros s sectional view of the maxilla. ................................ ................................ ................................ ............................... 25 3 1 Hypothetical proportion of adult body mass and brain mass achieved by UM 32773 at death compared to growth in Saimiri sciurus .. ................................ .............................. 47 3 2 The crania of specimens in the sample . ................................ ................................ ............. 48 3 3 The labeled virtual endocast of Notharctus tenebrosus (AMNH 127167) in 6 views ....... 49 3 4 The virtual endocast of Notharctus tenebrosus (AMNH 127167) in a translucent rendering of it cranium in 3 views ................................ ................................ ..................... 50 3 5 The virtual endocast of Notharctus tenebrosus (USNM V 23277) ................................ ... 51 3 6 The virtual endocast of Notharctus tenebrosus (USNM V 23278) ................................ ... 52 3 7 The virtual endocasts of all specimens in the sample ................................ ........................ 54 3 8 The labeled virtual endocast of Smilodectes gracilis (USNM V 17994) in 6 views ......... 55 3 9 The labeled virtual endocast of Smilodectes gracilis (UM 32773) in 6 views .................. 56 3 10 The virtual endocast o f Smilodectes gracilis ( USNM V 17996 ) ................................ ....... 57 3 11 The virtual endocast of the adult male Smilodectes gracilis (USNM V 17994) and the juvenile Smilodectes gracilis (UM 32773) in a transparent rendering of its sku ll ............ 59 3 12 The virtual endocast of Adapis parisiensis (NHM M 1345) ................................ .............. 60 3 13 The virtual endocast of Adapis parisiensis (NHM M 1345) in a translucent rendering of it cranium ................................ ................................ ................................ ....................... 61 3 14 A comparative plot of body mass estimates of the adapiforms calculated from different regression equations using dental proxies. ................................ .......................... 64 3 15 A comparative plot of body mass estimates of Notharc tus tenebrosus and Smilodectes gracilis calculated with all mammal regression equations of Gingerich (1990) using lower limb proxies. ................................ ................................ ....................... 67

PAGE 8

8 3 16 A comparative plot of body mass estimates of Nothar ctus tenebrosus and Smilodectes gracilis calculated with primates only regression equations of Boyer (2009) using lower limb proxies. ................................ ................................ ....................... 68 3 17 A comparative plot of body mass estimates of Nothar ctus tenebrosus and Smilodectes gracilis calculated with all mammal regression equations of Gingerich (1990) using upper limb proxies. ................................ ................................ ....................... 69 3 18 A comparative plot of body mass estimates of Notha rctus tenebrosus and Smilodectes gracilis calculated with primates only regression equations of Boyer (2009) using upper limb proxies. ................................ ................................ ....................... 70 3 19 quotients calculated from the body mass estimates of the adapiforms calculated u sing different dental eq uations . ........ 72 3 20 primates. ................................ ................................ ................................ ............................. 73 3 21 Convex polygon plot of olfactory bulb v olume versus endocranial volume ..................... 74 3 22 Convex polygon plot of olfactory bulb volume versus body mass. . ................................ .. 75

PAGE 9

9 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science THE FIRST VIRTUAL ENDOCASTS OF NORTH AMERICAN ADAPIFORM PRIMATES By Arianna R. Harrington May 2015 Chair: Jonathan I. Bloch Majo r: Zoology Notharctidae is an Eocene North American adapiform euprimate (crown primate) with a plentiful fossil record, making examination of their endocasts ideal for inferring the neuroanatomy and encephalization of stem strepsirrhines. Virtual endocast s of the notharctids Notharctus tenebrosus (n=3) and Smilodectes gracilis (n=3) from the Bridger formation of Wyoming and the late Eocene European adapid adapiform Adapis parisiensis (n=1) were reconstructed from high resolution X ray computed tomography ( CT) scan data. While the three species shared many neuroanatomical similarities differentiating them from plesiadapiforms (stem primates) and extant euprimates, cerebral proportions of N. tenebrosus were more variable than S. gracilis , which may be related level of cranial sexual dimorphism. One specimen of S. gracilis preserves the first record of a ? Sylv ian sulcus for this species. Dental body mass estimates suggest that the encephalization quoti ent (EQ) of N. tenebrosus was higher than S. gracilis , which is surprising as they are inferred to have shared common diets, activity patterns, and inferred social structures, and are purportedly closely related. However, body mass estimates using postcran ial variables suggest that some specimens of both species were of similar body size and EQ, suggesting that dental proxies overestimate the body mass of N. tenebrosus , which was likely about as encephalized as

PAGE 10

10 or slightly less encephalized than S. gracilis . R egardless, the results of this study suggest that the EQ of notharctids were not much higher than those of plesiadapiforms and most likely did not overlap those of extant strepsirrhines and haplorhines.

PAGE 11

11 CHAPTER 1 INTRODUCTION Taxonomic and Phylogenetic O verview extinct relatives (Szalay and Delson, 1979; Martin, 1990). Plesiadapiforms are a paraphyletic clade of stem primates from the Paleocene to Eocene that includes Plesiadapoidea, which is thought to be the sister group of Euprimates (Bloch et al., 2007). Omomyiformes (also known as Omomyoidea) and Adapiformes (also kn own as Adapoidea ) are two broad groups of e arly Euprimates. While omomyiforms are recognized from t he earliest Eocene to early Miocene of Asia, Europe , and North Am erica (Fleagle, 2013), Adapiformes is recognized from the earliest Eocene to late Miocene of Europe, Asia, Africa, and Asia. Adapiformes is typically divided into six families: the Eocene Eu ropean family Adapidae, the Eocene European, African, and Asian family Caenopithecidae, the Eocene through Miocene Asian family Sivaladapidae, the Eocene Asian and European family Cercamoniidae, and the Eocene European and predominantly North American fami ly Notharctidae (Rose, 1994; Gunnell, 2002; Maiolino et al., 2012; Fleagle, 2013). The taxonomic and phylogenetic groupings of adapiforms have been long discussed and debated, ory, 1920; Beard et al., 1986; Martin, 1990). However, some recent phylogenetic work spurred by studies of Darwinius massilae has somewhat controversially placed adapiforms with haplo rhines (Franzen et al., 2009) and even stem anthropoids (Gingerich et al. , 2010). While adapiforms and anthropoids share potential synapomorphies (Kay et al., 1997; Maiolino et al. 2012), these similarities have been suggested to be the result of convergent evolution (Seiffert et al., 2009). Results from other recent phylogenet ic analyses (Williams et al., 2010a; 2010b; Maiolino et al.,

PAGE 12

12 2012) have classified Adapiformes as stem strepsirrhines, while omomyoids have been classified as stem Haplorhines (Williams e t al., 2010b). Notharctus and Smilodectes are two genera of North A merican adapoid notharctids of the subfamily Notharctinae. Notharctus is present from the late Wasatchian (equivalent to the Wa 7 or Lostcabinian) biochron to the early Uintan (Ui 1; Unita A; Shoshonian) biochron of the North American Land Mammal Ages (NAL MA; Gunnell, 1995; Gunnell, 2002; Gunnell et al., 2008, Gunnell et al., 2009 ; see Figure 1 1 ). This correspond s to the late early Eocene to middle Eocene epoch , approximately 52.4 million years ago (Ma) to 45.7 Ma. Smilodectes is present from the early Br idgerian (Br1; Bridger A; Gardnerbuttian) biochron to the late Brigerian (Br3; Bridger C; Twinbuttian) biochron, corresponding to the early to mid Middle Eocene epoch, approximately 51 Ma to 46.3 Ma Ma (Robinson et al., 2004; Gunnell et al., 2008). Nothar ctus and Smilodectes are sometimes hypothesized to be sister taxa that evolved from the earlier notharctid genus Cantius (Covert, 1990), while others hypothesize a sister taxon relationship of Smilodectes and another notharctine genus, Copelemur , with Noth arctus grouping closer to a paraphyletic Cantius (Beard, 1988; Gunnell, 2002). Fossils of Notharctus and Smilodectes are relatively common and well known from the mammalian fossil record of North America, particularly in the middle Bridgeria n biochro b n (equivalent to Br 2, Bridger B, and the upper Blackforkian; Gun nell, 1998 and 2002; Gunnell et al . , 2008), which has been dated to be approximately 47.96 ± 0.13Ma (Murphey et al., 1999). Numerous postcrania and cranial samples of Notharctus and Smilodectes have been discussed in the literature (Gregory, 1920; Gazin, 1958; Hamrick and Alexander, 1996; Alexander and Burger, 2001, Maiolino et al., 2012; Boyer et al., 2013 b ). In particular, Notharctus tenebrosus and Smilodectes gracilis are two Nor th American notharctines known from several nearly

PAGE 13

13 complete skulls and skeletons, and have a substantial fossil record (Martin, 1990; Gunnell, 2002). Notharctus tenebrosus and S. gracilis were contemporary species that are found in middle Bridgerian deposi ts in the Black Forks Member of the Bridger Formation in the Green River Basin of Wyoming (Gunnell et al., 2008). Because of the relative abundance of specimens in good condition, specimens of N. tenebrosus and S. gracilis have been included in many types of analyses (Gunnell, 2002 ). This study takes advantage of this relative abundance of specimens and document s the first virtual endocasts of N. tenebrosus and S. gracilis reconstructed from the high resolution x ray computed tomography (CT) scan data of th e crania of N. tenebrosus and S. gracilis . Endocasts Endocasts are the casts of the internal cranial cavity (Jerison, 1973). The crania of small mammals are often capable of preserving fairly anatomically accurate molds of their contents because they have thinner meninges surrounding the brain than larger bodied mammals, allowing for their morphology to be precisely molded onto the endocranial surface (Jerison, 1973; Gurche, 1982; Macrini et al., 2007 ). Endocasts of fossil euprimates are particularly inter esting in light of the fact that a relatively enlarged brain to body size compared to other mammal lineages is often stated as a characteristic of the order Primates (Cartmill, 1992). However, how, when, and why the brains of primates began to expand durin g their evolutionary history has been a topic of discussion (Silcox et al., 2010). One method used to test hypotheses of morphologic evolution and pattern of encephalization in Primates has been by comparison of the endocasts of numerous fossil and extant primate taxa to each other and to other mammal groups. made of materials such as latex, and virtual endocasts reconstructed from high resolution x ray

PAGE 14

14 CT scan data. Nume rous natural and artificial endocasts of primates have been described and analyzed. These include the natural endocasts of the plesiadapiforms Microsyops annectens (Microsyopidae, Primates; Szalay, 1969; Silcox et al. , 2010) and Plesiadapis cookei (Plesiad apidae, Primates; Gingerich and Gunnel, 2005), the omomyiforms Necrolemur antiquus (Microchoeridae; Hürzeler, 1948; Hofer, 1962; Radinsky, 1970; Gurche, 1982) and Tetonius homunculus (Anaptomorphinae, Omomyoidae; Radinsky, 1967, 1970; Gurche, 1982), and th e adapiforms Adapis parisiensis (Adapinae; Adapidae; Neumayer, 1906; Gregory, 1920; Le Gros Clark, 1945; Radinsky, 1970; Gingerich and Martin, 1981 ; Gurche, 1982; Martin, 1990), and, in fact, S. gracilis and N. tenebrosus . The natural endocast of S. gracil is was first figured and described in detail by Gazin (1965), and data further elaborating on the endocasts of this species have appeared in Radinsky, (1970), Gingerich and Martin ( 1981 ), and Gurche (1978 and 1982). In addition, partial endocasts of N. ten ebrosus were first briefly described, but not figured, by Gregory (1920), and a partial endocast of N. tenebrosus and an olfactory bulb cast of Notharctus sp. were described and figured in Gurche (1978 and 1982). More recently, reconstructed virtual endoc asts have been utilized in several studies addressing early primate brain evolution. To date, virtual endocasts of primates include the plesiadapiforms Ignacius graybullianus (Paromomyidae, Primates; Silcox et al., 2009b), M. annectens (Silcox et al., 2010 ), Plesiadapis tricuspidens, and P. cookei (Orliac et al., 2014). A virtual endocast of TMM 40688 7, the holotype of the phylogenetically enigmatic euprimate from the Eocene of Texas, Rooneyia viejaensis is also known from Kirk et al. (2014), which supplem Hofer and Wilson (1967). These endocasts have demonstrated many of the advantages of studying endocasts digitally. Because the reconstruction of virtual endocasts from CT scans is

PAGE 15

15 not as potentially destructive as artificial endocast casting or the exposure of natural endocasts by peeling away the bones of the cranial vault, the use of this technology presents an unprecedented opportunity to visualize endocasts of speci mens where it would have been previously impossible. In addition, virtual endocast reconstruction allows for increased surface resolution relative to natural endocasts and makes more consistent comparisons possible, as empirical measurements such as volume s may be measured accurately and obtained consistently by the same programs. In past studies of early primate endocasts, brain volumes were estimated mathematically by methods such as double graphic integration (Jerison, 1973; Radinsky, 1977), infilling of cranial space by mustard seeds (Gingerich and Martin, 1981 ), and by water displacement (Gurche, 1978), and were compiled comparatively (Gurche, 1982; Martin, 1990) despite disparate methods of estimation. Similarly, while the relative brain size to body size is often compared among primate taxa using encephalization quotients (or EQ, the measure of the relative volume of a brain of an animal in question to the brain of a typical mammal of its size; Jerison, 1973), they were not often done using consistent body mass estimation methods. For example, the first body masses for S. gracilis, A. parisiensis, and N. tenebrosus approximated by comparing a composite skeleton of N. tenebrosus to those of the extant primat es. Jerison ( 1973) equated the body mass of N. tenebrosus to Lemur mongoz , while N. tenebrosus was likely closest in size to Propithecus verreauxi . Gurche (1978) used a similar method, comparing the reconstructed body length of a N. tenebrosus skeleton (AMNH 11478) to those of Hapalemur , Lemur , and Daubentonia . Jer ison, Radinsky, and Gurche all esti mated N. tenebrosus , S. gracilis,

PAGE 16

16 and A. parisiensis to be the same size, and combined these body m ass estimates with their independent volumetric estimates for brain size to arrive at EQ values for the respective species. More recent analyses (Martin, 1990; Gingerich and Gunnell, 2005; Silcox et al., 2009a; Silcox et al., 2010; Orliac et al., 2014; Ki rk et al., 2104) interested in calculating early primate masses to various measurements of the cranium and post cranial skeleton. These body mass equations includ e those calculated from measurements of the teeth (Gingerich et al., 1982; Conroy, 1987; Dagosto and Terranova, 1992), long bones (Gingerich, 1990), crania (Martin, 1990; Silcox et al., 2009a), and tarsals (Dagosto and Terranova, 1992; Boyer et al. 2013 a , Yapuncich et al., 2015). Despite some of these methodological disparities, olfactory bulb volumes, absolute brain volumes, and encephalization quotients estimated or obtained from these natural and artificial endocasts have been utilized in several synth etic studies of primate olfactory evolution (Takai et al., 2003 ; Heritage, 2014) and brain evolution (Silcox et al . , 2010; Steiper and Seiffert, 2012; Montgomery, 2010). However, it is clear that the past studies of early primate encephalization suffer fro m the lack of large sample sizes and comparably collected data. This study documents the largest sample of the first virtual endocasts of notharctine primates in order to rectify this situation. Ins titutional A bbreviations The following are the institutio nal abbreviations used in the text: AMNH, American Museum of Natural History, New York, NY ; NHM, Natural History Museum, London, United Kingdom; USNM, United States National Museum (Smithsonian Institution), Washington, D.C.; UWBM, University of Washingto n, Burke Museum of Natural History and Culture, Seattle, WA; YPM, Yale Peabody Museum, New Haven, CT.

PAGE 17

17 Figure 1 1. A graphic representation of the time periods in which the genera Notharctus and Smilodectes and the species Notharctus tenebrosus and Smilod ectes gracilis were present in North America. The first column identifies the epochs and the second column indentifies the biochrons in which these taxa were present. The third column identifies alternate names that each biochron is also known by. The red numbers represent time from present (in millions of years) bordering each biochron. Constructed using data from Gunnell (1995), Clyde et al. (1 997 ) ; Gunnell (1998); Murphey et al. (1999); Clyde et al. (2001); Gunnell (2002), Robinson et al. (2004), Gunnell et al. (2008), and Gunnell et al. (2009).

PAGE 18

18 CHAPTER 2 MATERIALS AND METHODS Virtual endocasts of a total of seven specimens were reconstructed in this study: three Notharctus tenebrosus (AMNH 127167, USNM V 23277, and USNM 23278), three Smilodectes gr acilis (UM 32773/MPM 2612, USNM V 17994, and USNM V 17996), and one Adapis parisiensis (N H M M 1345). AMNH 127167 is a partial skeleton of Notharctus tenebrosus . Axial elements include a nearly complete cranium, mandible, many cervical, thoracic, and lumba r vertebrae. Appendicular elements includ e the distal right and left femora , incomplete left and proximal right tibiae, an incomplete fibula, well preserve d right and left humeri and unl n a e, a left radius, and a nearly complete set of left hand elements (B oyer et al., 2013). The cranium of AMNH 127167 is somewhat dorso ventrally compressed and is slightly distorted. The lateral margin of the right supraorbital process is pulled posteriorly on the coronal plane relative to the left supraorbital process, and the right maxilla seems to be relatively compressed on the sagittal plane when compared to the left maxilla. Both supraorbital processes and the right maxilla bear puncture marks that have been suggested to have been inflicted by the claws of a predaceous bird at the time of death (Alexander and Burger, 2001). Based on relative canine size and flare of the cheek bones, AMNH 127167 has been inferred to be a male (Alexander, 1994; Alexander and Burger, 2001). The specimen was discovered in 1988 by Frank Ippol ito and Eugene S. Gaffney of the American Museum of Natural History at the locality of Butch Hill on the north side of the Fork member of the Bridger formation (H amrick and Alexander, 1996). Thus, this specimen is of the Br 2 biochron or the middle Bridgerian North American Land Mammal Age, approximately

PAGE 19

19 47.96 million years ago (Robinson et al., 2004). Several studies have already utilized both cranial and post crani al portions of AMNH127167 (Hamrick and Alexander, 1996; Alexander and Burger, 2001; Silcox et al., 2009a; Boyer et al., 2013 b ). USNM V 23277 is a specimen of Notharctus tenebrosus that was collected by Frank L. Pearce of the Smithsonian Institution in Augu st 1959, collected 10 miles east of Lyman, WY, which would most likely put the specimen in the Bridger B or Br2 biochron, similar to AMNH 127167, as this town is situated in an area with outcrops of that age (Gunnell, 1998). The cranium is rather complete, missing only portions of the right postorbital bar, zygomatic arch, right canine, and premaxilla. While left zygomatic arch shows some lateral crushing, the braincase seems relatively undistorted with no evidence of breakage along the dorsal or lateral su rface neurocranium. USNM V 23278 is also a specimen of Notharctus tenebrosus collected by Frank L. Pearce in August 1959. This specimen was noted to have been collected 0.5 miles south of Church Buttes, which is north east of Lyman, WY, and located in an a rea probably within the Bridger B formation, or the Br2 biochron (Gunnell, 1998). This specimen consists of a cranium and mandible, and the cranium shows more damage than the other two N. tenebrosus specimens examined for this study. While the specimen see ms to show minimal dorsoventral or lateral compression, it shows bilateral damage to the orbital and zygomatic regions and ventrally to the occipital, squamosal, petrosal, palatine, and maxillary regions with lack of preservation and damage of many teeth, including the right tooth row anterior to P 3 and the left tooth row anterior to M 1 . Both US NM V 23277 and USNM V 23278 have been evaluated to have medium tooth wear based on their premolars (Sauther and Cuozzo, 2012).

PAGE 20

20 UM 32773 (=MPM 2612) is an incomplete cranium of a juvenile Smilodectes gracilis from the University of Michigan locality known as BRW 15 in Wyoming, located in the lower Bridger formation and dated to Bridgerian biochron Br2, early middle Eocene. UM 32773 preserves a fairly comple te neurocra nium and basicranium al though the bullae are broken, portions of the frontals, squamosals, jugals, nasals, and premaxilla are missing, and there is lack of good preservation in the orbital region. The right maxilla is better preserved than the left. The cr anium is associated with a right dentary, but no postcrania. The subadult status of UM 32773 is denoted by mostly erupted M 3 , partially erupted P 2 , P 3 and P 4 in the crypt underneath the dP 3 and dP 4 , and canine in the crypt that may be observed in the right maxilla in CT scans ( Figure 2 1). The right dentary also preserves a dp 3 and dp 4 , erupting incisors and p 2 , and fully erupted m 1 , m 2 , and m 3 is also present, indicating th at the specimen was likely finished weaning, as the first molars are erupted (Smith, 1991). In addition, many sutures in the cranial vault are clearly unfused, including the metopic, coronal, sagittal, and parietal interparietal sutures. While many bones are missing, there does not appear to be any signs of dorso ventral crushing or later al distortion. This specimen has also been sampled in a numbe r of studies (Rose et al., 1999 ; Silcox et al., 2009a) USNM V 17994 and USNM V 17996 are both hypotype specimens of adult Smilodectes gracilis figured by Gazin (1958). USNM V 17994 was collected in 1940 by Frank L. Pearce in the northern portion of Twin Buttes in Sweetwater County, Wyoming. It consists of a nearly complete cranium missing only fragments of the right bulla, premaxilla, and I 1 P 1 . This specimen also includes axial and appendicular elements, including the calcaneus. The cranium is minimally distorted, exhibiting little signs of d orso ventral crushing and only a slight lateral shift of the orbits to the left.

PAGE 21

21 USNM V 17996, also collected by Frank L. Pearce, was found in a locality kno wn as Pinnacle Rock, in an ash layer 24 miles south east of the town of Granger, in Sweetwater County, Wyoming. This locality is found in Bridger B level and has also been described as being located between the town of Little America, Wyoming, and Twin But tes (Grande and Lundberg, 1988). This specimen is much less well preserved, consisting of a distorted cranium and incomplete mandibles. While the cranial vault is rather complete and appears undistorted, the rostral portion of this specimen is shifted left and is missing many parts, primarily from the orbital region, nasals, frontals, premaxilla, and maxilla anterior to P 3 on both sides. NH M M 1345 is a specimen of Adapis parisiensis that consists of a nearly complete cranium missing only the premaxilla an d portions of the left jugal and squamosal. This specimen has appeared numerous times in literature and its endocranial cast has in fact been described or discussed in numerous publications (Gingerich and Martin 1981 ). An artificial endocast produced from the endocranium of M 13 45 also has the catalog number NH M M 20192 and has been sampled in other studies before (Gurche 1978 and 1982). Some distortions in this specimen have been noted before, pri marily that there is some dorso ventral crushing in the front al bone that may have caused flattening of the frontal lobe region of the previously described endocast (Le Gros Clark, 1945; Gingerich and Martin, 1981 ). For this study, the cranium of each specimen was scanned using high resolution x ray computed tomogr aphy (CT). AMNH 127167 was scanned using the OMNI X Industrial Scanner at the Center of Quantitative Imaging (CQI) at Pennsylvania State University and UM 32773 was scanned using the Nikon XTH 225 ST scanner at the Nikon Metris X Tek Metrology Systems, Ltd Headquarters in Brighton, MI. The other five specimens were scanned using the Nikon XTH 225 ST scanner at the Shared Materials Instrumentation Facility at Duke University. The scanner

PAGE 22

22 settings, voxel size, number of slices, size of resulting tiff images, and size of the dataset for e ach scan is recorded in Table 2 1. The datasets produced from AMNH 127167, UM 32773/MP 2612, USNM V 17994, and NHM M 1345 were then manually segmented, a virtual endocast visualized, and linear and volumetric measurements w ere taken following p rotocols described by Silcox et al . (2010) using ImageJ (Rasband, 1997 2014) and the three dimensional visualization programs Avizo 6.6, Avizo 7.0, and Avizo 8.1.1 (Visuali zation Sciences Group, 2007 2015 ). USNM V 23277, USNM V 23278, and USNM V 17996 were segmented and visualized solely on Avizo 8.1.1 using the the same protocols as the others. The resulting endocasts and measurements were comp ared to each other and to relevant data published for extant strepsirrhines and haplorrhines and other fossil primates. All encephalization quotients (EQ) were calculated using equations from Eisenberg (1981; see Table 2 2 ). While an EQ equation from Jeri son (1973) has also been widely used, small or very large mammals (Martin, 1990). on. In order to aid the calculation of EQ estimates, body masses for the specimens were estimated using several methods. Because all of the specimens preserved an upper M 1 , the length and width of the M 1 from each specimen was measured to calculate its are a as in Gingerich et al. (1982). The length and width of the M 1 was also measured using protocol described in Gingerich et al. (1982) for each specimen as well, when available. If the specimen preserved both the right and left side of each tooth, then the average area of the right and left teeth were used. While

PAGE 23

23 Gingerich et al. (1982) reports an oft used body mass estimation equation using M 1 area ( Table 2 2) , this equation has come under some critique for failing to account for correction factors necessar y for conversion between logarithmic a nd numeric scales (Smith, 1993). For this study, body mass estimate calculated with a regression equation of the areas of M 1 and M 1 from Gingerich et al. (1982) was multiplied by the SE correction factor from Smith (19 93; Table 1). In addition, an M 1 and M 1 area body mass estimation regression equation calculated for strepsirrhin es with correction factors (Dagosto and Terranova, 1992) was used to estimate body masses of the fossil sample . However, because Notharctus te nebrosus has been noted to have larger molars than Smilodectes gracilis while sharing similar skull size (Gazin, 1958), postcranial variables were also measured to calculate body mass estimates to compare with those obtained from dental proxies. Diameters and lengths of the femur, tibia, humerus, and ulna were taken following protocols described in Alexander et al. (1979). Unfortunately, many of the specimens in the endocast sample are not associated with postcrania, so other dentally and postcranially asso ciated specimens of N. tenebrosus and S. gracilis were used as a comparative sample. The specimens included for postcranial analysis included AMNH 127167, the only specimen of N. tenebrosus in the endocast sample associated with postcrania; AMNH 11474, a r ather complete skeleton of N. tenebrosus figured in Gregory (1920); AMNH 11473, a specimen of N. tenebrosus with associated dentition and a humerus; AMNH 13024, a specimen of N. tenebrosus with associated dentition and a femur; and UWBM 81144, a skull and skeleton of Smilodectes gracilis . Because only one S. gracilis skeleton was available for study, data for species means were obtained for dental and postcranial measurements from Covert (1985) as a supplemental source.

PAGE 24

24 Two sets of equations equating meas urements of postcraninal variables to body mass estimates were used to calculate body masses of these specimens; the first was calculated from regressions sampling a suite of mammals (Gingerich, 1990) and the second was calculated from a p rimates only samp le (Boyer, 2009 ). Gingerich (1990) includes a code for calculation of body masses using lengths and diameters for long bones in the program Microsoft QuickBASIC; this code was re written as a script for the statistical program R (R Core Team, 2014), and ut ilized to calculate body mass estimates for this study; see Append ix B. Equations from Boyer (2009 ) were also written into a R script, and is reproduced in Appendix B. In addition, methods of measurement, calculation, and values for body mass, brain volum e, and EQ estimates from previous studies of N. tenebrosus , S. gracilis, and A. parisiensis specimens were compared to the methods and measurements that resulted in this study. Data from previous studies were collected from Radinsky (1977), Gurche (1978), Gurche (1982), Jerison (1979), Gingerich and Marti n (1981), and Martin (1990), and are reported in Table A 2. Body mass and brain volume data from extant taxa were obtained from Isler et al. (2008). These values are reported in Table A 3. In order to asses s the likely age, percent total adult weight and percent total brain mass achieved at time of death using comparative modern data for the juvenile S. gracilis , UM 32773, the tooth eruption sequence from this specimen was compared to those of living primate s (Tattersall and Shwartz 1974, Smith et al., 1994) and approximate ages at tooth eruption was determined by comparison to data from Saimiri sciurus (Long and Cooper, 1968). While AMNH 127167 and USNM 17994 have been suggested to be adult males of their re spective species (Alexander, 1994; Alexander and Burger, 2001), the sexes of the other adults in this sample have never been assessed. Where available, the length and width of canine roots

PAGE 25

25 were measured and craniofacial characters were assessed for USNM V 23277, USNM V 23278, USNM V 17996, and UM 32773 following the criteria in Alexander (1994). These characters were re assessed for AMNH 127167 and USNM 17994 for comparison. Resulting measurements were compared to data from A lexander (1994; see Table A 4). If a digital surface was available, a ll measurements were made by the 2 D measurement tool in Avizo 8.1. 1. Otherwise, digital calipers were used. Fi gure 2 1. A) Occlusal view of the cranium of UM32773 and B) a CT cross sectional view of the maxilla. The orange line on A denotes where the cross sectional view was taken.

PAGE 26

26 Table 2 1. High resolution CT scanner settings for each specimen in the sample. Specimen Species Vox. Vol. Amp. Proj. AMNH 127167 Notharctus tenebrosus .0626x.0626x.0662 180 20 1137 US NM V 23277 Notharctus tenebrosus .0397x.0397x.0397 135 55 2000 USNM V 23278 Notharctus tenebrosus .0382x.0382x.0382 135 55 2000 USNM V 17994 Smilodectes gracilis .0378x.0378x.0378 155 65 2000 USNM V 17996 Smilodectes gracilis .0348x.0348x.0348 135 55 20 00 UM 32773 Smilodectes gracilis .0324x.0324x.0324 NMH M 1345 Adapis parisiensis .0385x.0385x.0385 Table 1. CT scan settings of each specimen. UM 32773 does not have a CT scan setting because it was an early test run and the data has since been lost. Settings a ata for NHM M 1345 are also not available. Vox.= voxel dimensions in mm 3 ; Vol.= voltage in kV; Amp.= amperage in MicroA; Proj.= number of slices or projections . Table 2 2 . Summary of equations used in calculations in this study Sour ce Equation Eisenberg (1981) EQ = E/(0.055W 0.74 ) Gingerich et al. (1982) BM M 1 = e (2.72 + 1.62 (ln A)) * 1.076 Gingerich et al. (1982) BM m 1 = e (3.55 + 1.49 (ln B)) * 1.049 Dagosto and Terranova (1992) BM M 1 = e (2.04 + 1.62 (ln A)) * 1.12 Dagosto a nd Terranova (1992) BM m 1 = e (2.87 + 1.57 (ln B)) * 1.07 A= upper first molar area; B = lower first molar area; BM M 1 = body mass estimate calculated from upper first molar area; BM m 1 = body mass estimate calculated from lower first molar area; e = approxi number; E= brain volume in cm 3 ; EQ= encephal ization quotient; ln= natural log; W=b ody mass in grams (g) . Correction factors from Smith (1993) and Dagosto and Terranova (1992) are already added to each body mass equation.

PAGE 27

27 CHAP TER 3 RESULTS Developmental Stage of UM 32773 The Smilodectes gracilis UM 32773 died as a juvenile, with a dP 3 /dP 3 , dP 4 /dP 4 , fully erupted M 1 /M 1 , M 2 /M 2 , P 1 , incompletely erupted M 3 /M 3 , erupting P 2 /P 2 , and unerupted upper canine, P 3 , and P 4 ( Figure 2 1). W hile the dental eruption sequence for S. gracilis is unknown, if the dental eruption pattern can be reasonably assumed to be similar to that of N. tenebrosus upper and lower eruption (Gregory 1920; Gingerich and Smith, 2010) , the most likely dental eruptio n sequence of the teeth of S. gracilis is M P M M P P P .With comparison to the available data, there is no single modern primate that closely models this upper tooth eruption sequence, although it is perhaps most c omparable to that of of Eulemur mongoz (M I I C 1 M M P C 1 P P ; Tattersall and Schwartz, 1974).Unfortunately, there is limited data available relating eruption of the teeth to age of the juvenile, body mass, o r brain mass of the juvenile at different ages, for Eulemur mongoz . Franzen et al. (2009) used dental eruption sequence and several measures of growth in Saimiri sciurus to model the likely age and brain and body weight of the juvenile holotype of Darwiniu s masillae ( Ida ; comprised of the specimens PMO 214.214 and WDC MG 210) proportional to its adult brain and body mass at the time of death. Compared to the D. masillae holotype, the upper teeth of UM 32773 are similar in having an erupting canine, dP 3 and dP 4 with P 3 and P 4 forming underneath, and a fully erupted M 1 . There is some evidence that UM 32773 was more mature, however, as the M 2 is still erupting and the M 3 is erupting without roots in Ida while the M 2 is fully erupt ed and M 3 is erupting with roots in UM 32773. However, the P of Ida are fully erupted, which differs from the erupting condition

PAGE 28

28 of the P of UM 32773, suggesting that the sequence of premolar eruption was different between S. gracilis and D. ma ssilae . While the sequence of eruption of the maxillary teeth of S. sciurus (M 1 M 2 I 1 I 2 (P 4 P 3 ) P 2 M 3 C) and the mandibular teeth (M 1 M 2 I 1 I 2 M 3 P 4 P 2 P 3 C) are not synchronous (Long and Cooper, 1968), the eruption sequence of the right dentary of UM 32 773 was compared to the mandibular eruption sequence of S. sciurus and D. massilae , as was in Franzen et al., 2009 ( Figure 3 1). If the dental eruption sequences and growth trajectories of S. gracilis and S. sciurus are comparable, it suggests that the b rain mass of UM 32773 was just over 98% of its adult brain mass, and that the body mass of UM 32773 was approximately 63 64% of its adult body mass. Sex of USNM V 23277, USNM V 23278, and USNM V 17996 Available upper canine measurements for each specimen are recorded in Table 3 1. While the upper canine measurements alone suggest that USNM V 23277 and USNM V 23278 are female compared to previously reported canine dimensions ( Table A 1) , both specimen also possess diminished sagittal crests and rounded, non triangular posteri or edge of the coronal suture ( Figure 3 2) which are cranial features of female N. tenebrosus. In addition, while the zygomatic arch of USNM V 23277 is slightly crushed laterally, the face appears narrower and 127167 and more similar to the female specimen of AMNH 11466 (Alexander, 1994; Gregory, 1920). While USNM V 17996 does not male S. gracilis USNM V 17994, USNM V 17996 does not possess any i ndication of a sagittal crest ( Figure 3 2). Furthermore, the temporal lines on the surface of the frontal and parietal bone do not meet mid sagittally until almost at the lambdoidal suture, a charac ter of female S.gracilis (Alexander, 1994). The sex of UM 32773 could not be assessed because 1) the adult canines

PAGE 29

29 were unerupted at the time of death, 2) the zygomatic arches are ill preserved, 3) the temporal lines are ill defined, and 4) the lack of a s agittal crest could be attributed to the specimen not having developed that cranial secondary sexual characteristics at the time of death. Description of the Endocasts of Notharctus tenebrosus The virtual endocasts of Notharctus tenebrosus may be describe d by three specimens. All three endocasts preserve some portion of the anterior most part of the brain, the olfactory bulbs, followed by a circular fissure that separate them from a non overlapping cerebrum consisting of ill defined frontal, temporal, and occipital lobes, which also do not overlap th e cerebellum following it. AMNH 127167 ( Figure s 3 3 and 3 4) is the most complete of the three N. tenebrosus endocasts, with only minor damage to the anterior most olfactory bulbs. USNM V 23277 is missing som e of the anterior and ventral olfactory bulbs and has poor preservation of ventral paleocortex ( Figure 3 5). USNM V 23278 has a better preserved dorsal and ventral cerebral surface than USNM 23277, but lacks portions of the cerebellum in addition to some a nterior and ventral po rtions of the olfactory bulbs ( Figure 3 6). Overall, the three virtual endocasts of N. tenebrosus are variable in general outline due to differing proportions of the cerebrum, the biggest difference being the ratios between the maximu m width and maximum len gth ( Table 3 2 ). Overall, USNM V 23277 is more elongate in overall proportions than USNM V 23277, which is more elongate in appearance than AMNH 127167. Maximum endocast height/length ratio is less variable, and maximum height/width ratio of AMNH 127167 is less than the identical ratios of the other two specimens. The total endocranial volumes for the specimens are as follows: 7.35 cm 3 (AMNH 127167), 8.11cm 3 (USNM V 23277), and 7.46 cm 3 (USNM V 23278). Due to the preservation, it cann ot be determined whether any of the virtual olfactory bulbs of N. tenebrosus

PAGE 30

30 Nonetheless, details like the numerous nerve roots leading to the cribiform plate in the ethmoid bone may be obs erved on the olfactory bulbs of AMNH 127167. Overall, the olfactory bulbs comprise about 15% of the total length and 2.11% of the total volume of the en docast of AMNH 127167 ( Table 3 2 ). The anterior most portion of the olfactory bulbs in all the N. tenebr osus specimens in the sample seem to terminate past the posterior edge of the postorbital bars ( Figure s 3 4 , 3 5, and 3 6 ). The cerebrum is divided into several parts. Roughly, the frontal lobes may be approximated as the portion of the cerebellum anterio r to the S ylvian sulcus. The temporal lobes are roughly pos terior to the S ylvian fissure, and merge posteriorly into the occipital lobes. AMNH 127167 has short frontal lobes and a wide temporal lobe, while USNM V 23277 has longer frontal lobes and a nar rower temporal pole. This can be demonstrated by the relative maximum cerebral width to length rations: AMNH 127167 has the shortest cerebrum relative to width at 1.28, while USNM V 23277 has the most elongate cerebrum at 0.95. USNM V 23278 is intermediate between t he other two, at 1.05 ( Table 3 2 ). All three endocasts preserve bilateral casts of the orbitotemporal canal, a vascular structure that connects the contents of the postglenoid foramen to the orbits at the craniorbital foramen (Kirk et al . , 2014 ). The orbitotemporal canal runs rostro caudally on the lateral side of the temporal lobe, and is generally accepted to denote the position of the rhinal sulcus based on comparisons to the brains of extant primates and mammals (Gazin, 1965; Radinsky, 1974; Gingerich and Martin, 1981 ). The position of the orbitotemporal canal (and thus the rhinal fissure) is low on the lateral aspect of all three endocasts, dividing the cerebrum into the neocortex dorsally and the paleocortex ( or the pyriform cortex or lobes ) ventrally. The pyriform

PAGE 31

31 lobes of all three notharctines are oval shaped from ventral view, although the less well preserved pyriform lobes of USNM V 23277 are more elongate in appearance than the other two. On the dorsal surface of the virtual cerebra, the superior sagittal sinus runs postero anteriorly along the mid sagittal plane as a raised ridge on the dorsal surface. On all three N. tenebrosus specimens, the superior sagittal sinus is best defined just anterior to the confluence of the sagittal and transverse sinuses, becoming less defined anteriorly and undefined by about half the length of the cerebrum. It is worth noting that in this study the transverse sinus is defined differently from that of Gazin (1965) and Gingerich and Martin ( 1981 ), who re ferred to the sigmoid sinus (in the sense of Gurche, 1978, Silcox et al., 2009 b , Silcox et. al., 2010, and Kirk, 2014) interchangeably as the lateral sinus and the transverse sinus. In this study, the transverse sinus is defined as a sinus that often trans ects the cerebrum or cerebellum that joins the superior sagittal sinus medially and the petrosquamous sinus laterally. On each side of the superior sagittal sinus is the lateral sulcus, which forms well defined valleys in all three specimens, although most distinctly on USNM V 23278. Shallow bilateral depressions on the dorsal outline of the anterior cerebrum are identifiable as Sylvian sulci , fissure, or fossae. Ventrally, the depression of the Sylvian is better defined and intersects the rhinal fissure, overlain by the orbitotemporal canal. Dorsally, the Sylvian sulcus is not well defined as a groove, but all three specimens show a depression. More variable in presence is the ?suprasylvian sulcus tentatively identified on AMNH 127167; it is a shallow sulcus on the dorsal surface of the cerebrum that parallels the outline of the edge of the temporal lobe on each side the endocast. This feature is better defined on the left side of AMNH 127167 than the right, and is absent from the endocasts of USNM V 23277 and USNM V 23278. While AMNH 127167

PAGE 32

32 does not preserve meningeal vessels on the cerebrum, the other two N. tenebrosus specimens possess several branching meningeal vessels that sweep back caudally on the temporal lobes. There is no midbrain expo sure on any of the virtual endocasts of N. tenebrosus . The cerebellum lies posterior to and is not overlapped by the cerebrum. In each, the cerebellum is divided into the vermis and two lateral lobes by the paramedian fissures. On the endocast of AMNH 1 27167, however, there appears to be a cast of a vessel overlying the paramedian fissure, distinctly on the right side and perhaps less distinctly on the left side. The tranverse sinus marks the division between the cerebrum and cerebellum. The transverse s inus is well defined in all three specimens, and each converges with the petrosquamous sinus and sigmoid sinus. The petrosquamous sinus is fully enclosed in a bony canal in all three specimens, and connects the transverse sinus with the contents of the pos tglenoid foramen (Kirk, 2014). The sigmoid sinus wraps around the posterior and lateral edges of the lateral lobes of the cerebellum, and is contiguous with the contents of the jugular foramen; its contents also exit out of a foramen on the parietal squamo sal suture (Gingerich and Martin, 1981 ). The paraflocculus lies laterally to each lateral lobe of the cerebellum, ventral to the sigmoid sinus. The size of all paraflocculi in the three specimens seem comparable. A ?fissura prima is tentatively identifiabl e on the vermis of each specimen. The cast of the pituitary (hypophyseal) fossa is located on the ventral surface of the endocasts. Length, width, and height of the structure are recorded in Table 3 2 . In overall shape, the pituitary is round in AMNH 1 27167, but is longer than wide in an anterior posterior direction in USNM V 23277 and USNM V 23278. The midpoint of the cast of the pituitary lies about 5.71 mm caudal in AMMNH 127167 and 5.28 mm caudal in USNM V 23278 to the rostral edge of the optic c hiasm, but the anterior edge of the optic chiasm is too damaged to assess in USNM V

PAGE 33

33 23277. The pituitary cannot be seen from the lateral aspect of USNM V 23277 and USNM V 23278, but can be seen in AMNH 127167. A structure apparently homologous to the pr omontory ramus of the internal carotid artery identified on Smilodectes gracilis (Gazin, 1965) may be seen on the endocasts just posterior to the pituitary on either side, although that of USNM V 23278 appears damaged on the right side. Just posterior to t he internal carotid artery is the inferior petrosal sinus, which runs bilaterally in a posterior to antero medial direction. This region is the most complete on AMNH 127167 and is damaged on the right side in USNM V 23277 and USNM V 23278. The cast of the promontorial artery can be seen on the ventral surface of USNM V 23277 as a J shaped structure extending roughly dorsal to the anterior end of the inferior petrosal sinus and terminating just posterior to the cast of the mandibular branch of the trigeminal nerve (V 3 ). This structure is not observed in AMNH 127167 and is only preserved on the left side of the endocast o USNM V 23278. Major cranial nerves are well preserved on the ventral surface of all three endocasts, including the optic nerve (II) stemmi ng from the optic chiasm and leading to the optic canal. This structure is slightly distorted in the endocasts of AMNH 127167 and USNM V 23277 and is most symmetrical in USNM V 23278. Just caudal to the optic nerve are the casts of the contents of the s phenorbital fissure (or the foramen lacerum anterius in Gazin, 1965): the oculomotor nerve (III), the trochlear nerve (IV), the ophthalmic branch of the trigeminal nerve (V 1 ), the abducens nerve (VI), and the ophthalmic vein. The cast of the content of the foramen rotundum, the maxillary branch of the trigeminal nerve (V 2 ), follows immediately caudally. These three structures are clearly defined and visible on the ventral surface of the three endocasts. Posterior to the pituitary, the mandibular branch (V 3 ) of the trigeminal nerve is represented as the cast of the content of the foramen ovale. Casts of the contents of the internal auditory meatus, the facial

PAGE 34

34 nerve (VII) and vestibulocochlear nerve (VIII), are present as clearly forked structures slightly ant erior and ventral to the paraflocculi of all three endocasts. Posterior to the paraflocculi are the casts of the contents of the jugular foramen (or foramen lacerum posterius): the glossopharyngeal nerve (IX), the vagus nerve (X), accessory nerve (XI), and the jugular vein. The hypoglossal nerve (XII) to the hypoglossal foramen may be seen as a slight outgrowth of the brain stem, and it is not well preserved on any of the endocasts of N. tenebrosus . They are identifiable bilaterally on the endocast of AMNH 127167 but only on the left side of USNM V 23277 and USNM V 23278. Comparisons to Previously Published Endocasts of Notharctus tenebrosus Endocast anatomy of of Notharctus was briefly characterized, but not illustrated, by Gregory (1920), who based his descriptions on an incomplete and natural endocast of AMNH 11478 ( Notharctus tenebrosus; then Notharctus tyrannus ) and partial skulls of AMNH 14656 ( Cantius venticolus; then Notharctus venticolus ), AMNH 12569 ( Notharctus tenebrosus ; then Notharctus osborni ), and AMNH 13030 ( Smilodectes gracilis ; then Notharctus matthewi ). He noted an endocast shape with generally small cerebrum and frontal lobes and observed that the cerebrum did not overlap the cerebellum nor olfactory bulbs, which he surmised were separat ed surface of the frontal bones of one of the specimens. He also notes that the endocast of Notharctus possesses a superior longitudinal sinus (= superior sagi ttal sinus) flanked by the sulcus intraparietalis (=lateral sulcus) on either side. Gazin (1965) also briefly discussed t he partial endocast o f AMNH 11478 as a comparison to the endocasts of S. gracilis , noting that while the surface is weathered, the endo cast seems to exhibit a weak ?suprasylvian sulcus on the right dorsal surface, and estimated the total endocast length (excluding the olfactory bulbs) was about 32.5mm to 35 mm long, with a temporal lobe width between 27.5mm and 30 mm. This is

PAGE 35

35 comparable t o the length and width measurements of the virtual endocasts of N. tenebrosus reconstructed in this study ( Table 3 2) , although that of USNM 23277 is longer antero posteriorly (37.8 mm) and narrower temporally (25.5 mm) than the estimated range of values f or AMNH 11478. Gurche (1978) expanded these brief descriptions of the endocast morphology of Notharctus d Gurche, 1982), casts of petrosal bones of AMNH 12585, AMNH 12575, AMNH 12581, AMNH 12569, and YPM 12958, and olfactory bulb casts of USNM 244365 and one of an unknown number ctory bulb cast of Notharctus referred to by R adinsky (1970) in a discussion of the olfactory bulb anatomy of S. gracilis . Gurche (1978) also discusses the olfactory bulb morphology of USNM 23278 from an x ray photograph in the dorso ventral aspect, noting that the olfactory bulbs of the specimen USNM 23278, which suggests that the olfactory area of the specimen is too damaged to reconstruct a complete enough olfa ctory bulb for the specimen that could show diverging bifurcation in the olfactory lobes. In general, Gurche (1978) agrees with the findings of Gregory (1920), although disagrees with Gazin (1965) on the identification of the ?suprasylvian sulcus on the pa rtial natural endocast of AMNH 11478, due to the irregular nature of the weathered endocast surface. Gurche (1978) also notes that the ?suprasylvian sulcus is absent in endocast A. This data and the current study seems to support that the presence of the ? suprasylvian sulcus is variable in the endocasts of N. tenebrosus . In other features, with the exceptions already noted, the previous descriptions of the morphological features of the endocasts of N. tenebrosus and

PAGE 36

36 Notharctus agree with the data presented in this current study. The current study adds a plethora of new information about the ventral surface, including information on the cranial nerves, cast of the pituitary, and endocranial vasculature which was not available in previously known natural endoc asts of Notharctus . An endocranial volume for N. tenebrosus was estimated by Gurche (1978, 1982) using a olfactory bulbs) and a minimal portion of the left cerebellu m, and was reported to be 10.43 cm 3 , although also reconstructed be around a minimum of 8.91 cm 3 and a maximum of 13.29 cm 3 . These values are higher than the absolute volumes of the reconstructed virtual endocasts in this study, with even the minimum recon structed volume surpassing the virtual endocast of N. tenebrosus with the largest absolute volume (USNM V 23277; 8.06 cm 3 ). This seeming overestimation of volume may be due to the fact that the natural endocasts of S. gracilis was the reconstruction (Gurche, 1978), and the current study suggests that the general endocast shape of N. tenebrosus appears much more variable intraspecifically and is not necessarily similar in shape to the endocasts of S. gracilis in all individuals ( Figu re 3 7). mm 3 (with a minimum and maximum restoration of 178.2 mm 3 an d 375.2 mm 3 ) using the technique of water displacement (Gurche, 1978). It is larger than the estimate of 155.5 mm 3 for the virtual endocast of AMNH 127167, which has the most complete olfactory bulb of the N. tenebrosus individual with a larger absolute brain volume. However, given that a large amount of

PAGE 37

37 probably unreliable. Descripti on of the Virtual Endocasts of Smilodectes gracilis The virtual endocasts of USNM V 17994 ( Figure 3 8), UM 32773 ( Figure 3 9 ), and USNM V 17996 ( Figure 3 10 ) are similar in characteristics to one another, with consistently heart shaped cerebra with a taper ing frontal region, expanded temporal pole, and a rounded occipital region. None of the endocasts have cerebra that overlap the cerebellum or olfactory bulbs. The total volumes of the endocasts are as follows: 8.42 cm 3 (UM 32773), 8.63 cm 3 (USNM 17994), an d 8.99 cm 3 (USNM 17996). Relative proportions of all three are fairly similar. With the olfactory bulbs excluded, the maximum width to length ratio fall between 0.89 and 0.92, maximum height to length between 0.50 and 0.53, and maximum height to width betw een 0.55 and 0.60 ( Table 3 3 ). The virtual endocasts of USNM V 17994, USNM V 17996, and UM 32773 preserve most of their olfactory bulbs. The olfactory bulbs of USNM V 17994 is actually complete; they are bifurcate, show many small olfactory nerves branchi ng off into foramina in the cribiform, and are separated from the cerebrum by a shallow circular fissure that is better defined on the lateral aspects than the dorsal or ventral aspect ( Figure s 3 8 and 3 9 ). While the olfactory bulbs of USNM V 17996 and UM 32773 are less well preserved ventrally and anteriorly, they may evince roughly bifurcate outlines. The anterior most olfactory bulbs appear to terminate posterior to the postorbital bar in all three s pecimens ( Figure s 3 10 and 3 11 ). The olfactory bulb v olume of USNM V 17994 is 177.7mm 3 or about 2.06% of the total endocast volume. On the cerebrum of all three specimens, the orbitotemporal canal is low on the lateral sides, dividing it into the neocortex and the paleocortex (pyriform cortex or lobe). The p yriform lobes are roughly oval in shape. The orbitotemporal canal is contiguous with the postglenoid

PAGE 38

38 vein, which exits out of the postglenoid foramen, and the petrosquamous sinus. While the petrosquamous sinus is enclosed in a bony canal from its exit at t he parieto squamosal foramen in the two adult S. gracilis virtual endocasts, the separation of the petrosquamous sinus is not quite definite in the juvenile UM 32773. In this specimen, there is intermittent contact of the petrosquamosal sinus to the occipi tal surface of the cerebrum. Other vascular structures are more similar among the virtual endocasts, although the inferior petrosal sinuses of USNM 17996 are not well defined on the ventral surface. The principle cerebral features that may be seen on the endocast surface are 1) an ill defined superior sagittal sinus that is only very apparent in USNM V 17994, 2) shallow lateral sulci running in parallel to the superior sagittal sinus, 3) the preservation of suture lines from the frontal and parietal bones on the endocast of UM 32773 and 4) the apparent lack of sylvian sulci on UM 32773 and USNM V 17996. USNM V 17994, however, preserves a shallow sulcus anterior and lateral to the lateral sulcus on each side of cerebrum that can tentatively be identified as a Sylvian sulcus. This ?Sylvian sulcus can be better identified on the left side of USNM 17994, and the right side has a ?vessel in this same region. ?Vessels also seem present on the right side of UM 32773 and the left side of USNM V 17996 where a sylvian sulcus would be expected. The appearance of other meningeal vessels on the cerebrum also seems variable; UM 32773 has several smaller branching vessels on the temporal lobes and a larger ?vessel on the right cerebrum where a ?suprasylvian sulcus may be ex pected. None of the S. gracilis endocasts in this sample have preserved ?suprasylvian sulci. USNM V17996 has numerous smaller branching meningeal vessels on the temporal and occipital lobes. Finally, USNM V 17994 has a minimal meningeal vasculature. While the cerebrum and cerebellum do appear demarcated on all endocasts, the transverse sinuses between them are not well defined in any of the virtual

PAGE 39

39 endocasts. However, all three endocasts seem to possess a relatively well defined raised area at the confluenc e of the superior sagittal sinus and the transverse sinuses. None of the S. gracilis specimen show midbrain exposure. The cerebellum is composed centrally of a vermis flanked by lateral lobes on either side, separated by paramedian fissures. None of the endocasts show raised vessels overlaying the paramedian fissure. A ?fissura prima can be identified on the vermis of all three S. gracilis endocasts. The sigmoid sinus is present on the lateral margins of the lateral lobes of all specimens, confluent with the petrosquamous sinus. The sigmoid sinus wraps around dorsally and slightly posterior to the paraflocculus, which are round and flank either side of the lateral lobes. On the ventral surface, the overall shape of the cast of the pituitary is consisten tly round in all three specimens, with hypohysis length to width ratio s between 1.01 and 1.23 ( Table 3 3 ). In the two adult specimens, the pituitary is tall enough to be seen when the endocast is viewed laterally. The cranial nerves of the three virtual en docasts are consistent, preserving the optic nerves, optic chiasm, and separate roots for the contents of the sphenoidal fissure and foramen rotundum. The optic nerve and optic chiasm is the most damaged on UM 32773. The mandibular branch of the trigemina l nerve is consistently parallel to what is identified as the promontorial branch of the internal carotid artery, with the cast of the facial (VII) and auditory (VIII) nerve in a position posterior to it. In USNM V 17996 and UM 32773, the forked appearance of nerve VII and VIII can be observed well, but the preservation detail is poorer in USNM V 17994. All possess the contents of the jugular foramen posterior to the contents of the internal auditory meatus, with casts of the hypoglossal foramen on the brai n stem. The hypoglossal nerve is best defined on UM 32773 and less well defined on USNM V 17994 and USNM V 17996.

PAGE 40

40 Comparisons to Previously Published Endocasts of Smilodectes gracilis The first natural endocasts of Smilodectes gracilis were described in d etail by Gazin (1965), based on the specimens USNM 23276, a fairly complete natural endocast missing only the olfactory bulbs, and USNM 17997, a skull with a partially exposed endocast in the cerebellar portion. The former specimen was recovered during fie ldwork near Cedar Mountain in the Bridger Basin of Wyoming in 1959, and was reportedly found in the Bridger B layer. Radinsky (1970) compared USNM 23276 and a newly prepared specimen, YPM 12152, in an additional, brief description of the endocasts of this species. Gurche (1978) reviewed this literature and added additional information from a fragmentary skull of UKMNH 4050 and the petrosal of AMNH 13030, a specimen originally attributed to Notharctus discussed by Gregory (1920). In general, the virtual endo casts of USNM V 1799 4, USNM V 17996, and UM 32773 ( Figure 3 7) are similar in characteristics to the endocast morphology of USNM 23276. The biggest difference between the previously described endocasts and the virtual endocasts is the addition of data from the olfactory bulbs, cranial nerves, and vascular organization, as well as empirical Before this study, the olfactory bulbs of S. gracilis were poorly known, as US NM 23276 did not preserve this feature (Gazin, 1965). Radinsky (1970) prepared YPM 12152 to expose the right olfactory bulb, and using a cast of an olfactory chamber assigned to Notharctus (later illustrated the olfactory bulb of S. gracilis to be widely divergent and terminating posterior to the postorbital bar. The current study confirms this finding, although the degree of divergence of the olfactory bulbs illustrated by Radinsky (1970) may be s lightly exaggerated. The ventral preservation of USNM 23276 was not as clear as that of the virtual endocasts, and thus had no identifiable differentiation between the contents of the sphenorbital fissure

PAGE 41

41 (cranial nerves III, IV, V, VI, and the ophthalmic vein) and foramen rotundum (cranial nerve V 2 ; Gazin, 1965), although it was noted that the contents of the respective foramina should be separated due to their separate preservation on the cranium of S. gracilis specimens (Gurche, 1978). Indeed, the bette r resolution of the virtual endocasts demonstrates that there was indeed a separation of the contents of the foramen rotundum and the sphenorbital fissure on the endocasts of S. gracilis . Otherwise, the cranial nerve organization found in this study is con sistent with that found by Gazin (1965) and Gurche (1978). Neither the postglenoid vein nor the petrosquamous sinuses were preserved in the previously known natural endocasts, as USNM 23276 is only exposed on the left side for that part of the endocast. O n the exposed left side, those vessels as well as the cast of the paraflocculus have evidently been broken off, as they are not illustrated in Gazin (1965), Radinky (1970), nor Gurche, (1978 and 1982). Measurements previously made from the natural endocas ts of S. gracilis are relatively comparable to the data obtained from the virtural endocasts in this study. Gazin (1965) estimated the length of the cerebrum and cerebellum of USNM 23276 to be about 32.5mm, which i s only larger by about 1mm than the same m easurement in the virtual endocasts ( Table 3 3 ). The estimate of 24 mm for the length of the cerebrum is also similar to the measurements of the virtual endocasts, as is the estimate of 29.7 mm for the maximum width of the endocast across the cerebrum of U SNM 23276. Combined, these measurements give a similar cerebral width to length ratio for the natural and virtual endocasts, ranging from 1.24 to 1.27, signifying a consistently wider than long cerebral shape. Volumes of S. gracilis endocasts have been es timated for both USNM 23276 and YPM 12152. Jerison (1973) first estimated total endocast volume using the double graphic integration

PAGE 42

42 method on the composite illustration of USNM 23276 and YPM 12152 from Radinsky (1970), which yielded a volume of 9.60 cm 3 . Double graphic integration is a method by which endocast volume is approximated to the volume of an elliptical cylinder. The height of this cylinder is the length of the endocast from the anterior most cerebrum (without the olfactory bulbs) to the position of the hypoglossal nerve. The two axes of the elliptical cylinder are equal to the average height of the endocast in lateral view and average width of the endocast from dorsal view (Jerison, 1973). This is a particularly handy method because endocast volu mes may be been questioned (Radinsky, 1977). Radinsky (1977) used a method of comparing double graphic integration and water displacement of hand made clay models o f a composite Smilodectes endocast to arrive at a volume estimate of 9.50 cm 3 . Gurche (1978) also used water displacement on clay reconstructions to estimate total endocranial volumes (including the olfactory bulbs) for USNM 23276 (9.12 cm 3 , with a likely minimum and maximum volume of restoration between 8.84 cm 3 and 9.26cm 3 ) and YPM 12152 (9.95 cm 3 with a likely range of 8.84 cm 3 12.52 cm 3 ). Together, these most likely volume estimates for S. gracilis are from between 0.2 cm 3 to 1.32 cm 3 larger than the e stimates obtained for the virtual endocasts of adult S. gracilis, depending on the estimate . While the estimates of Jerison (1973) and Radinsky (1970) are probably unreliable and overestimated because of the methods in which they were obtained, the endocas t volume of USNM 23276 obtained by Gurche (1978) is believable. As a mostly revealed endocast, USNM 23276 probably required a minimal amount of reconstruction with clay compared to YPM 12152, which is described as a specimen with only half of the dorsal su rface of the endocast exposed (Gurche, 1978; Radinsy, 1970). Clearly, there is some evidence that the level of

PAGE 43

43 exposure and completeness of the endocasts available for study influenced how close the volume estimates were to the volume measurements obtained from the virtual endocasts. The Virtual Endocast of Adapis parisiensis As mentioned previously, the endocast of NHM M 1345 and morphology of the endocasts of A. parisiensis has been described in some detail already (LeGros Clark, 1945; Gingerich and Mart in, 1981 ). Overall, the virtual end ocast of NHM M 1345 ( Figure 3 12 ) is fairly complete, and for the first time visualizes the anatomy of the petrosquamous sinus and postglenoid vein. The petrosquamous sinus was completely encased in at least two bony cana ls, and it connects the contents of the postglenoid foramen with contents of the canal to the foramen on the parieto squamosal suture and the transvers and sigmoid sinus. The petrosquamous sinus is well separated from the narrow occipital lobe of the cereb ru m by thick bone (see Figure 3 13 , dorsal view). There are also some details of the meningeal vasculature preserved on the virtual endocast. The total volume of the virtual endocast is 8.81 cm 3 ( Table 3 4 ), which is minutely larger than the volume previ ously reported for the latex endocast of NHM M 1345, 8.80 cm 3 (Gingerich and Martin, 1981 ), and the volume of 8.31 cm 3 reported by Gurche (1978 and 1982) of BM. M. 20192, another artificial endocast of NHM M 1345. However, this is probably accounted for b y the evidence that there was some shrinkage during the casting of the latex endocast (Gingerich and Martin, 1980). The olfactory bulbs are 211.50 mm 3 , or 2.40% of the total volume. This is lower than a previous olfactory bulb volume for this specimen (266 .6 cm 3 , with a probable range from 264.5 cm 3 to 334.4 cm 3 ) estimated by Gurche (1982). Body Mass Estimates and Encephalization Quotients Dental lengths and widths are reported in Table 3 1 and the resulting body mass estimates calculated for each speci men are summarized in Table 3 5 and illustrated in Figure 3 14 . In

PAGE 44

44 general, dental equations from Gingerich et al. (1982) yield higher body mass estimates than the equations from Dagosto and Terranova (1992). In both sets of equations, the M 1 equation yields a higher body mass than the M 1 equations. Furthermore, t he body masses estimates for N. tenebrosus specimens are higher than those of S. gracilis . In fact, there is no overlap in body mass for N. tenebrosus and S. gracilis when dental proxies are used, alt hough the 95% confidence intervals of the species mean of N. tenebrosus and S. gracilis overlap in all dental proxies . Body masses calculated for USNM 17996, AMNH 127167, and USNM 23277 using the prosimian mandibular M 1 equation are lower than those calculated using their equation for the upper M 1 , but the opposite is true for USNM 23278. The modified upper M 1 equation of Gingerich yields noticeably higher body masses than Dagosto and prosimian upper M 1 equations, with USNM 23277 yielding the highest body mass at 5576g, which is nearly double the estimate for the species using the other equation, at 2943 g. The estimated body mass of A. parisiensis , NHM M 1345, is lower than the other two adapiforms in the sample. Measurements of the postcrania are recorded in Table 3 6. When postcranial proxies are used to calculate body mass, mixed results occur ( Table s 3 7 and 3 8 and Figure s 3 15, 3 16, 3 17, and 3 18). In general, body masses calculated using the all mammal r egressions from Gingerich (1990) are higher than those estimated using the primates o nly regressions from Boyer (2009 ). Notharctus tenebrosus and Smilodectes gracilis show overlapping body masses when femoral diameter, femoral length, and humeral length a re used as proxies. When tibial diameter, tibial length, and humeral diameter are used, Smilodectes gracilis yields higher body mass estimates. When ulnal length is used, Notharctus tenebrosus yields a higher body mass.

PAGE 45

45 Eisenberg encephalization quotients obtained for specimens in this study are reported in Table 3 9 and comparat ively illustrated in Figure 3 19 . The lowest for N. tenebrosus specimens are obtained using the body masses from the modified upper M 1 equation of Gingerich (0.26 0.28), and 1 (0.40 0.44) and M 1 (0.40 S. gracilis by dental estimates are consistently higher than those of N. tenebrosus using the same methods (0.41 1 equation , 0.66 0.81 and 0.76 0.95 using Dagosto and 1 and M 1 equations, respectively). The EQ of A. parisiensis is the highest amongst the other adapiforms in the sample when using equations of the upper M 1 . E ncephalization quotient values of the fossil adapiforms calculated from dental proxies compared to plesiadapiforms, Rooneyia viejaenis, and extant strepsirrhines a nd haplorhines are reported in Table A 2 and illustrated in Figure 3 20 . Encephalization quot ients of 1 range from 0.26 ( P. tricuspidens) to 0.48 ( I. graybullianus ). These values overlap with EQ values for N. tenebrosus and S. gracilis calculated using the same body mass proxies. Some EQ estimates for S. gracilis and A. parisiensis strepsirrhines (0.86 2.46), while showing no overlap with the range for extant haplorhines (1.11 N. tenebr osus using all body mass estimation methods (0.26 0.67) do not overlap with those of any extant taxa in the sample. This is true even when postcranial proxies are used for the N. tenebrosus specimen AMNH 127167, which yields an EQ range of (0.32 0.72) with body masses calculated using humeral diameter, humeral length, ulnal diameter, and ulnal length. The EQ of Rooneyia viejaensis obtained using the Gingerich M 1 body mass estimates fall within the rage of the adapiforms calculated for this study, but the EQ calculated

PAGE 46

46 from BM estimated from cranial length (Silcox et al., 2010) pushes it far out of the range the adapiforms and into the range of extant strepsirrhines and extant haplorhines. Relative Sizes of the Olfactory Bulbs The new olfactory bulb results are figure d in a co mparative context in Figure 3 21 . The olfactory bulb volumes relative to total endocranial volume of N. tenebrosus , S. gracilis , plot within the minimum convex polygon of extant strepsirrhines. Unlike the data presented in Kirk et al., 2 014, the new virtual endocast of A. parisiensis also plots within the minimum convex polygon of ex tant strepsirrhines. Figure 3 21 suggests that compared to the plesiadapiforms, the adapiforms in this sample had relatively small olfactory bulb volume relat ive to total endocranial volume. However, the adapiforms had larger olfactory bulb size relative to endocranial volume when compared to R. viejaensis . When olfactory bulb volume is compared to body mass, the pattern remains that the adapiforms in the sampl e plot within the minimum convex polygon of ex tant strepsirrhines. Figure 3 22 suggests that A. parisiensis had larger olfactory bulbs relative to body size when compared to the notharctine adapiforms in the sample. The plesiadapiforms in this sample had l arger olfactory bulbs relative to body mass or about the same olfactory bulb to body mass ratios compared to adapiforms in the sample. However, R. viejaensis had smaller olfactory bulbs relative to body mas s compared to both adapiforms and plesiadapiforms.

PAGE 47

47 Figure 3 1. Hypothetical proportion of adult body mass and brain mass achieved by UM 32773 at death compared to growth in Saimiri sciurus . Dental eruption and body weight data from Long and Cooper (1968), brain weight data from Manocha (1979), and da ta for Table 3 1. Dental measurements of specimens in the sample. Specimen Species CL CW CH M 1 L M 1 W M 1 L M 1 W AMNH 127167 N. tenebrosus 4.29 3.11 8.34 5.58 6.15 5.33 3.97 USNM V 23277 N. tenebrosus 3.49 2.64 5.56 6.57 5.37 4.12 USNM V 23278 N. tenebrosus 3.55 2.42 5.08 6.11 5.73 4.07 UM 32773 S. gracilis 4.64 5.34 4.55 3.26 USNM V 17994 S. gracilis 3.62 2.30 4.40 5.32 USNM V 17996 S. gracilis 4.19 5.40 4.24 3.33 NHM M 1345 A. parisiensis 4.38 4.50 AMNH 11474 N. tenebrosus 5.79 6.11 5.16 3.86 AMNH 11473 N. tenebrosus 5.44 4.04 AMNH 13024 N. tenebrosus 5.45 6.71 4.05 5.69 UWBM 88114 S. gracilis 4.38 4.70 3.22 4.40 Species mean S. gracilis 4.11 5.20 4.21 4.30 CL = canine root length; CW = canine root width; CH = canine height; M 1 L = length of the upper first molar; M 1 W = width of the upper first molar; M 1 L = length of the lower first molar; M 1 W = wi dth of the lower first molar. Species mean of S. gracilis from Covert (1985), with from a sample size n = 37 for M 1 and n =12 for M 1 measurements. All measurements in millimeters (mm).

PAGE 48

48 Figure 3 2. The crania of A) the adult male Notharctus tenebrosus AMNH 127167, B) the adult female N. tenebrosus USNM V 23277, C) the adult female N. tenebrosus USNM V 23278, D) the adult male Smilodectes gracilis USNM V 17994, E) the adult female S. gracilis USNM V 17996, and F) the juvenile S. gracilis UM 32773 in dor sal view. Note the degree of the flare of the zygomatics between A and B, the degree of development of the sagittal crest between D and E, and the unfused sutures on the cranial vault of F. Scale = 5mm.

PAGE 49

49 Figure 3 3. The virtual endocast of Notharctus te nebrosus (AMNH 127167) in A) ventral, B) dorsal, C) anterior, D) posterior, E) right lateral, and F) left lateral views. Scale = 5mm.

PAGE 50

50 Figure 3 4. The virtual endocast of Notharctus tenebrosus (AMNH 127167) in a translucent rendering of it cranium in A) right lateral, B) ventral, and C) dorsal views. Scale = 5mm.

PAGE 51

51 Figure 3 5.The virtual endocast of Notharctus tenebrosus USNM V 23277 in D) posterior, E) anterior, F) right lateral, G) left lateral, H) dorsal, and I) ventral views and in a transpa rent rendering of its skull in A) ventral, B) dorsal, and C) right lateral views. Scale = 5mm.

PAGE 52

52 Figure 3 6.The virtual endocast of Notharctus tenebrosus USNM V 23278 in D) posterior, E) anterior, F) right lateral, G) left lateral, H) dorsal, and I) ven tral views and in a transparent rendering of its skull in A) ventral, B) dorsal, and C) right lateral views. Scale = 5mm.

PAGE 53

53 Table 3 2. Measurements of the virtual endocasts of Notharctus tenebrosus. AMNH 127167 USNM 23277 USNM 23278 Total endocast length 40.95 42.66 38.11 Maximum endocast width 29.66 25.45 25.17 Maximum endocast height 15.66 17.73 17.66 Olfactory bulb length 6.17 4.82 3.29 Olfactory bulb width 8.97 8.17 8.28 Length of the endocast without olfactory bulbs 34.78 37.84 34.82 Olfactory bulb length/total endocast length 0.15 0.11 0.09 Endocast width/total endocast length 0.72 0.60 0.66 Endocast width/length of the endocast without OB 0.85 0.67 0.72 Endocast height/total endocast length 0.38 0.42 0.46 Endocast height/length of the end ocast without OB 0.45 0.47 0.51 Endocast height/width ratio 0.53 0.70 0.70 Maximum length of cerebrum 23.14 25.90 24.56 Endocast width/cerebral length ratio 1.28 1.05 0.95 Total endocast volume 7375.61 8060.26 7427.58 Volume of olfactory bulbs 155.45 180.20 112.18 Endocast volume without olfactory bulbs 7220.16 7880.06 7315.40 % of endocast composed of olfactory bulbs 2.11 2.24 1.51 Maximum width of cerebellar portion 20.04 17.57 Maximum cerebellar width/maximum endocast width 0.68 0.69 Hypo physis length 3.17 4.22 4.46 Hypophysis width 4.73 2.49 2.91 Hypophysis depth 1.01 1.84 1.32 Hypophysis length/hypophysis width ratio 0.67 1.69 1.53 Distance of midpoint of hypophysis to anterior edge of optic chiasm 5.71 5.28 Lengths are in millim eters (mm) and volumes are in mm 3 . OB= olfactory bulbs. Length of cerebrum was measured in the dorsal view from the posterior edge of the transverse sinus where it intersects the superior sagittal sinus to the posterior edge of the circular fissure. Total endocast volume does not include the volume of the postglenoid vein. Maximum width of the cerebellar portion does not include the sigmoid sinus.

PAGE 54

54 Figure 3 7. The virtual endocasts of Notharctus tenebrosus , Adapis parisiensis , and Smilodectes gracilis in A) right lateral, B) ventral, and C) dorsal views. Scale = 5mm.

PAGE 55

55 Figure 3 8. The virtual endocast of Smilodectes gracilis (USNM V 17994) in A) ventral, B) dorsal, C) anterior, D) posterior, E) right lateral, and F) left lateral views. Scale = 5mm.

PAGE 56

56 Figure 3 9. The virtual endocast of Smilodectes gracilis (UM 32773) in A) ventral, B) dorsal, C) anterior, D) posterior, E) right lateral, and F) left lateral views. Scale = 5mm.

PAGE 57

57 Figure 3 10 The virtual endocast of Smilodectes gracilis USNM V 17996 in D) posterior, E) anterior, F) right lateral, G) left lateral, H) dorsal, and I) ventral views and in a transparent rendering of its skull in A) ventral, B) dorsal, and C) right lateral views. Scale = 5mm.

PAGE 58

58 Table 3 3. Measurements of the virtual endocasts of S milodectes gracilis UM 32773 USNM 17994 USNM 17996 Total endocast length 36.75 40.21 40.71 Maximum endocast width 30.45 30.83 30.26 Maximum endocast height 17.36 16.86 18.16 Olfactory bulb length 3.80 6.32 6.69 Olfactory bulb width 9.19 9.11 9. 07 Length of the endocast without olfactory bulbs 32.95 33.89 34.02 Olfactory bulb length/total endocast length 0.10 0.16 0.16 Endocast width/total endocast length 0.83 0.77 0.74 Endocast width/length of the endocast without OB 0.92 0.91 0.89 Endocas t height/total endocast length 0.47 0.42 0.45 Endocast height/length of the endocast without OB 0.53 0.50 0.53 Endocast height/width ratio 0.57 0.55 0.60 Maximum length of cerebrum 23.97 24.51 24.34 Endocast width/cerebral length ratio 1.27 1.26 1.24 Total endocast volume 8420.47 8630.00 8990.74 Volume of olfactory bulbs 130.23 177.72 150.19 Endocast volume without olfactory bulbs 8290.24 8452.28 8840.55 % of endocast composed of olfactory bulbs 1.55 2.06 1.67 Maximum width of cerebellar portion 18.66 17.73 18.12 Maximum cerebellar width/maximum endocast width 0.61 0.58 0.60 Hypophysis length 3.48 4.00 4.51 Hypophysis width 3.44 3.75 3.68 Hypophysis depth 0.57 1.72 1.65 Hypophysis length/hypophysis width ratio 1.01 1.07 1.23 Distance of mid point of hypophysis to anterior edge of optic chiasm 6.04 7.20 7.91 Lengths are in millimeters (mm) and volumes are in mm 3 . OB= olfactory bulbs. Length of cerebrum was measured in the dorsal view from the posterior edge of the transverse sinus where it in tersects the superior sagittal sinus to the posterior edge of the circular fissure. Total endocast volume does not include the volume of the postglenoid vein. Maximum width of the cerebellar portion does not include the sigmoid sinus.

PAGE 59

59 Figure 3 11. The virtual endocast of the adult male Smilodectes gracilis (USNM V 17994) in a transparent rendering of its skull in A) right lateral B) dorsal and C) ventral views, and the virtual endocast of the juvenile Smilodectes gracilis (UM 32773) in a transparent ren dering of its skull in D) right lateral E) dorsal and F) ventral views. Scale = 5mm.

PAGE 60

60 Figure 3 12. The virtual endocast of Adapis parisiensis (NHM M 1345) in A) ventral, B) dorsal, C) anterior, D) posterior, E) right lateral, and F) left lateral views. Scale = 5mm.

PAGE 61

61 Figure 3 13. The virtual endocast of Adapis parisiensis (NHM M 1345) in a translucent rendering of it cranium in A) right lateral, B) ventral, and C) dorsal views. Scale = 5mm.

PAGE 62

62 Table 3 4. Measurements of the virtual endocasts of Adapis par isiensis. NHM M 1345 Total endocast length 45.68 Maximum endocast width 36.60 Maximum endocast height 19.01 Olfactory bulb length 9.09 Olfactory bulb width 7.54 Length of the endocast without olfactory bulbs 36.60 Olfactory bulb length/total endoc ast length 0.20 Endocast width/total endocast length 0.63 Endocast width/length of the endocast without OB 0.78 Endocast height/total endocast length 0.42 Endocast height/length of the endocast without OB 0.52 Endocast height/width ratio 0.66 Maximu m length of cerebrum 26.72 Endocast width/cerebral length ratio 1.07 Total endocast volume 8809.68 Volume of olfactory bulbs 211.50 Endocast volume without olfactory bulbs 8598.18 % of endocast composed of olfactory bulbs 2.40 Maximum width of cerebe llar portion 15.73 Maximum cerebellar width/maximum endocast width 0.55 Hypophysis length 5.90 Hypophysis width 4.80 Hypophysis depth 0.72 Hypophysis length/hypophysis width ratio 1.23 Distance of midpoint of hypophysis to anterior edge of optic c hiasm 10.83 Lengths are in millimeters (mm) and volumes are in mm 3 . OB= olfactory bulbs. Length of cerebrum was measured in the dorsal view from the posterior edge of the transverse sinus where it intersects the superior sagittal sinus to the posterior ed ge of the circular fissure. Total endocast volume does not include the volume of the postglenoid vein. Maximum width of the cerebellar portion does not include the sigmoid sinus.

PAGE 63

63 Table 3 5. Body mass estimates for specimens in the sample using dental pro xies. Specimen Species *M 1 M 1 *M 1 M 1 AMNH 127167 N. tenebrosus 5019 2463 3434 2265 USNM V 23277 N. tenebrosus 5554 2943 3684 2439 USNM V 23278 N. tenebrosus 4266 2271 3985 2650 UM 32773 S. gracilis 2943 1565 2030 1302 USNM V 17994 S. gracilis 276 6 1466 USNM V 17996 S. gracilis 2479 1309 1886 1205 NHM M 1345 A. parisiensis 2044 1078 AMNH 11474 N. tenebrosus 5273 2780 3151 2068 AMNH 11473 N. tenebrosus 3648 2414 AMNH 13024 N. tenebrosus 5563 2934 3915 2600 UWBM 88114 S. gracilis 2193 1157 1897 1212 Species mean S. gracilis 2330 1229 1842 1174 Body mass estimates calculated from different regression equations using dental proxies. *M 1 = an all mammal upper first molar area regression equation from Gingerich et al., 1982; M 1 = an all strepsirrhine upper first molar area equation from Dagosto and Terranova, 1992; *M 1 = an all mammal lower first molar area equation from Gingerich et al., 1982; M 1 = an all strepsirrhine lower first molar area equation from Dagosto and Terranova, 1992. A ll body masses calculated for this study have correction factors applied. S. gracilis species mean = species mean reported from Covert (1985). All masses in grams (g). Equations in table A 1.

PAGE 64

64 Figure 3 14. A comparative plot of body mass estimates of th e adapiforms Notharctus tenebrosus, Smilodectes gracilis, and Adapis parisiensis calculated from different regression equations using dental proxies. *M 1 = all mammal upper M 1 equation from Gingerich et al. (1982); *M 1 = all mammal lower M 1 equation from Gingerich et al. (1982); M 1 = strepsirrhines only upper M 1 equation from Dagosto and Terranova (1992); M 1 = strepsirrhines only lower M 1 equation from Dagosto and Terranova (1992). Data from T able 3 5. Equations in Table 2 2.

PAGE 65

65 Table 3 6. Measurements o f postcranial variables of specimens in the sample. Specimen Species FD TD HD UD FL TL HL UL AMNH 127167 N. tenebrosus 7.53 3.92 79.83 88.68 USNM V 23277 N. tenebrosus USNM V 23278 N. tenebrosus UM 32773 S. gr acilis USNM V 17994 S. gracilis USNM V 17996 S. gracilis NHM M 1345 A. parisiensis AMNH 11474 N. tenebrosus 7.89 5.41 6.26 123.70 105.26 72.65 AMNH 11473 N. tenebrosus 6.84 75.9 AMNH 13024 N. tenebrosus 8.05 138.08 UWBM 88114 S. gracilis 7.99 8.61 8.80 4.73 123.42 106.86 76.4 77.52 Species mean S. gracilis 7.4 (n=2) 8.7 (n=3) 3.2 (n=3) 132.8 (n=1) 109.5 (n=1) 74.6 (n=3) 85.6 (n=1) FD = femoral diameter; FL = femoral length; HD = humeral diameter; HL = humeral length; TD = tibial diameter; TL = tibial length UD = ulnal diameter; UL = ulnal length. S. gracilis species means from Covert (1985). All measurements in millimeters (mm). Table 3 7. Body mass estim ates of specimens in the sample based on equations from Gingerich (1990). Specimen Species FD TD HD UD FL TL HL UL AMNH 127167 N. tenebrosus 3050 3394 2785 USNM V 23277 N. tenebrosus USNM V 23278 N. tenebrosus UM 32773 S. gracilis USNM V 17994 S. gracilis USNM V 17996 S. gracilis NHM M 1345 A. parisiensis AMNH 11474 N. tenebrosus 3333 1445 1888 6352 3118 2638 AMNH 11473 N. tenebrosus 2376 2966 AMNH 13024 N. tenebrosus 3521 8505 UWBM 88114 S. gracilis 3450 5027 4573 6314 3265 3018 1936 Species mean S. gracilis 2796 5169 7669 3518 2832 2530 FD = femoral diameter; FL = femoral length; HD = humeral diameter; HL = humeral length; TD = tibial diameter; TL = tibial length UD = ulnal diameter; UL = ulnal length. S. gracilis species means from Covert (1985). All body masses in grams (g). Code used to calculate these body masses are reproduced in Appendix B.

PAGE 66

66 Tabl e 3 8. Body mass estimates of specimens in the sample bas ed on equations from Boyer (2009 ). Specimen Species FD TD HD UD FL TL HL UL AMNH 127167 N. tenebrosus 2679 1138 1147 1525 USNM V 23277 N. tenebrosus USNM V 23278 N. te nebrosus UM 32773 S. gracilis USNM V 17994 S. gracilis USNM V 17996 S. gracilis NHM M 1345 A. parisiensis AMNH 11474 N. tenebrosus 2446 2408 1688 2602 1864 1160 AMNH 11473 N. tenebrosus 2107 2185 AMNH 13024 N. tenebrosus 2578 3589 UWBM 88114 S. gracilis 2528 7093 3955 1762 2585 1956 1305 1110 Species mean S. gracilis 2069 7267 710 3203 2114 1235 1403 FD = femoral diameter; FL = femoral len gth; HD = humeral diameter; HL = humeral length; TD = tibial diameter; TL = tibial length UD = ulnal diameter; UL = ulnal length. S. gracilis species means from Covert (1985). All body masses in grams (g). Code used to calculate these body masses are repro duced in Appendix B.

PAGE 67

67 Figure 3 15. A comparative plot of body mass estimates of Notharctus tenebrosus and Smilodectes gracilis calculated with all mammal regression equations of Gingerich (1990) using lower limb proxies. M 1 = strepsirrhines only upper M 1 equation from Dagosto and Terranova (1992); Fem D. = femoral diameter; Tib. D = tibial diameter; Fem. L = femoral length; Tib. L = tibial length. Data from Table 3 7. Equations in Table 2 2.

PAGE 68

68 Figure 3 16. A comparative plot of body mass estimates of Notharctus tenebrosus and Smilodectes gracilis calculated with primates only regression equations of Boyer (2009) using lower limb proxies. M 1 = strepsirrhines only upper M 1 equation from Dagosto and Terranova (1992); Fem D. = femoral diameter; Tib. D = t ibial diameter; Fem. L = femoral length; Tib. L = tibial length. Data from Table 3 8. Equations in Table 2 2.

PAGE 69

69 Figure 3 17. A comparative plot of body mass estimates of Notharctus tenebrosus and Smilodectes gracilis calculated with all mammal regressio n equations of Gingerich (1990) using upper limb proxies. M 1 = strepsirrhines only upper M 1 equation from Dagosto and Terranova (1992); Hum. D. = humeral diameter; Hum. L. = humeral length; Uln. L. = ulnal length. Data from Table 3 7. Equations in Table 2 2.

PAGE 70

70 Figure 3 18. A comparative plot of body mass estimates of Notharctus tenebrosus and Smilodectes gracilis calculated with primates only regression equations of Boyer (2009) using upper limb proxies. M 1 = strepsirrhines only upper M 1 equation from Da gosto and Terranova (1992); Hum. D. = humeral diameter; Uln. D. = ulnal diameter; Hum. L. = humeral length; Uln. L. = ulnal length. Data from Table 3 8. Equations in Table 2 2.

PAGE 71

71 Table 3 9 . Endocast volume and encephalization quotients for specimens in th e sample. Specimen Species ECV *M 1 M 1 *M 1 M 1 AMNH 127167 N. tenebrosus 7.38 0.24 0.39 0.32 0.44 USNM V 23277 N. tenebrosus 8.06 0.25 0.40 0.34 0.46 USNM V 23278 N. tenebrosus 7.43 0.28 0.45 0.29 0.40 UM 32773 S. gracilis 8.45 0.47 0.67 0.55 0.76 U SNM V 17994 S. gracilis 8.63 0.45 0.71 USNM V 17996 S. gracilis 8.99 0.44 0.81 0.62 0.81 NHM M 1345 A. parisiensis 8.81 0.57 0.92 Encephalization quotients calculated using body mass estimates reported in Table 3 2, using the same legend. ECV= e ndocranial volume in cm 3 . All encephalization quotients (EQ) calculated using the EQ equation of Eisenberg (1982; see Table A 1).

PAGE 72

72 Figure 3 body mass estimates of the a dapiforms Notharctus tenebrosus, Smilodectes gracilis, and Adapis parisiensis calculated from different equations utilizing dental proxies. *M 1 = all mammal upper M 1 equation from Gingerich et al. (1982); *M 1 = all mammal lower M 1 equation from Gingerich et al. (1982); M 1 = strepsirrhines only upper M 1 equation from Dagosto and Terranova (1992); M 1 = strepsirrhines only lower M 1 equation from Dagosto and Terra nova (1992). Data from Table 3 8. Equations in Table 2 2.

PAGE 73

73 Figure 3 20. A comparative plot of primates. Data for N. tenebrosus, S. gracilis, and A. parisiensis from Table 3 9. Data for all other taxa from Table s A 2 and A 3.

PAGE 74

74 Figure 3 21. Convex polygon plot of olfactory bulb volum e versus endocranial volume. Data in Table A 4.

PAGE 75

75 Figure 3 22. Convex polygon plot of olfactory bulb volume versus body mass. Data in Table A 4.

PAGE 76

76 CHAPTER 4 DISCUSSION Overall, the endocasts of N. tenebrosus and S. gracilis are similar to one another. Bo th notharctine species have endocasts similar to that of the adapine A. parisiensis in having small frontal lobes and no cerebral overlap of the olfactory bulbs or cerebellum, which is also the condition found in the plesiadapiforms, and is presumably ance stral (Silcox et al., 2009; Silcox et al., 2010; Orliac et al . , 2014). This differs from the derived condition found in modern primates and the Eocene fossil primate Rooneyia viejaensis , which have cerebra that overlap both the olfactory bulb and cerebellu m (Martin, 1990; Kirk et al., 2014), and the fossil omomyiforms Necrolemur antiquus and Tetonius homunculus , which have cerebra that do overlap the cerebellum but not the olfactory bulbs. Cerebral overlap is thought to signify expansion of brain volume rel ative to the size of the cranium (sources), suggesting that the brains of adapiforms in general were not as relatively large as those of modern primates. However, unlike the plesiadapiforms, the notharctine enodocasts have relatively wide, globular tempora l lobes and relatively smaller olfactory bulbs. The notharctines are consistent with other euprimates in having no midbrain exposure at the juncture of the occipital lobe of the cerebrum and the cerebellum, although whether the presence of this characteris tic in many plesiadapiform endocasts is primitive or secondarily derived is uncertain (Edinger, 1964; Silcox et al., 2010). The widest part of the endocast in dorsal view of a notharctine endocast is about midway across the temporal region of the brain, wh ereas for plesiadapiforms, the widest line across may be drawn closer to the occipital lobes. In addition, the position of the rhinal sulcus is lower in the notharctine endocasts relative to the plesiadapiforms, similar to A. parisiensis, R. viejaensis , an d the omomyiform endocasts, suggesting that the neocortex of notharctines was more developed than that of plesiadapiforms (Silcox et al., 2009b; Silcox et al . , 2010; Orliac et

PAGE 77

77 al., 2014; Kirk et al., 2014). These details suggest that the brains of notharct ines, like other known early euprimate endocasts, are relatively more expanded and perhaps may have been more visually oriented than plesiadapiforms (Silcox et al., 2010). While many features of the endocast of S. gracilis and N. tenebrosus are similar in traspecifically and interspecifically, several differences are worth noting. Sylvian S ulcus Smilodectes gracilis is particularly noted in the primate brain literature for being the only known fossil or living primate that does not preserve the Sylvian sulc us (or Sylvian fissure; Gazin, 1965; Gurche, 1978; Radinsky, 1970; Martin, 1990; Silcox, 2010). This is significant as the Sylvian sulcus, along with the calcarine sulcus found in the internal wall of the occipital portion of the cerebrum, has been said to be a definitive characteristic of primate brains (Elliot Smith, 1902; Martin, 1990). The virtual endocast of M 1345, A. parisiensis , shows a typical primate Sylvian sulcus, which Elliot Smith (1902) defined as arising from an intersection with the rhinal sulcus (presumably under the orbitotemporal canal in these endocasts). The A. parisiensis endocast shows a groove at the Sylvian sulcus which is comparable to that of modern primates and the Eocene primates Rooneyia viejaensis, Tetonius homumculus , and Nec rolemur antiquus. On the other hand, the endocasts of N. tenebrosus show a depression on the ventral and the lateral sides of the cerebrum at the intersection of the rhinal sulcus where a Sylvian sulcus would be expected, but it is not as defined as in the endocast of A. parisiensis . A shallow groove does present itself ventrally, but does not continue as a defined groove on the dorso lateral surface. Thus, it is perhaps more appropriate to call the depression in the N. tenebrosus endocast a Sylvian fossa r ather than strictly a sulcus or fissure. The virtual endocasts of S. gracilis do not possess any semblance of a groove where a Sylvian sulcus would be expected, and with the exception of USNM 17994, there is almost no indication of even a Sylvian fossa.

PAGE 78

78 T he lack of the Sylvian sulcus in the endocasts of S. gracilis has been a source of much speculation. The presence of fissures on the brain is to a large extent affected by absolute brain volume. It has been noted that many small bodied mammals lack sulci o n the surface of their brain, especially with brain masses under 5 grams (Radinsky, 1975; Macrini et al., 2007). This is because the brains of mammals are arranged such that the cell bodies of neurons (grey matter) cover the surface of the brain while the inner volume consists of the axons of neurons (white matter). As volume increases, surface area increases more slowly; thus, in order for the volume of grey matter on the surface to keep in proportion to the volume of the inner white matter, the surface ar ea of the brain is increased by the introduction of sulci as the brain enlarges (Martin, 1990). However, it is unlikely that small absolute brain volume is the only reason why S. gracilis may not possess the feature, for the brains of all fossil and modern primates with brain volumes absolutely smaller than that of the S. gracilis , including those of Tetonius homunculus, Rooneyia viejaensis, Adapis , Tarsius , and even Microcebus murinus , with an average brain volume of only 1.63 cm 3 (Isler et al., 2007), all possess a S ylvian sulcus (Radinsky, 1970). The level of compaction of the brain within the cranium also affects the formation of sulci; thus, skull shape is another factor that can impact surface brain morphology (Martin, 1990 ). Radinsky (1970) suggested that the Sylvian sulcus could be absent because S. gracilis was a larger than fossil primates such as Tetonius . To some extent, it is evident that S. gracilis does have a lessened degree of postorbital constriction relative to A. parisiensis , which does have a more well defined Sylvian sulcus. However, N. tenebrosus appears to have a similar level of postorbital constriction as S. gracilis , but preserves a Sylvian sulcus . Alternatively, it has been suggested that the Sylvian sulcus was not absent at all on the brain of S. gracilis; its absence

PAGE 79

79 could be the result of thick meninges or subdural vessels overlaying the are a of the Sylvian sulcus, which c ould be hindering its preservation (Radinsky, 1970; Gurche, 1982). Gurche (1978) noted that the endocast of USNM V 23276 has faint, distinct ridges that could be traces of vessels over the area of the Sylvian sulcus; Radinsky (1970) does not identify any such ridges in USNM V 2 study provide some support for thi s hypothesis; all three of the virtual S. gracil is sample appear to show faint ridges around where a Sylvian sulcus would be expected on at least one side of the endocast. However, it is not certain whether subdural vessels alone could account for the lack of the Sylvian sulcus in S. gracilis . While it is not evident on the virtual endocast of AMNH 127167, the other two N. tenebrosus endocasts seem to show ridges on and near the area of the Sylvian sulcus. The ridges in the area of the Sylvian are thicker on USNM V 23277 and are finer and more numerou s on USNM V 23278. Given the relatively shallow Sylvian sulcus on the N. tenebrosus and its lack altogether on the majority of S. gracilis endocasts and the evidence for subdural vessels herein presented, it is possible that the presence of obscuring vascu lar structures over the Sylvian sulcus is a characteristic of notharctine endocasts. Regardless, even if the Sylvian sulcus was truly absent on the endocasts of S. gracilis , what significance it represents is uncertain. One possibility is that S. gracilis retained the ancestral condition, as plesiadapiforms apparently do not possess a clear Sylvian sulcus either; a Sylvian fossa is documented on the endocasts of Ignacius graybullianus (Silcox et al., 2009 b) and Plesiadapis tricuspidens (Orliac et al., 2014 ), but not on Microsyops annectens (Silcox et al., 2010). This could imply that the Sylvian sulcus evolved more than once in separate lineages of

PAGE 80

80 euprimates; or alternatively, that the notharctines diverged from the lineage before the Sylvian sulcus evolve d (Martin, 1990). Suprasylvian S ulcus Radinsky (1970) suggested that the presence of the ?suprasylvian sulcus on some endocasts of S. gracilis (e.g. USNM 23276) and the absence of the Sylvian sulcus for the species are most likely related. He pro posed th at as the orbital space intrudes more on the brain and promote the formation of the Sylvian sulcus, the resulting deformation could obscure the ?suprasylvian sulcus. Interestingly, no other Eocene euprimate endocasts appear to preserve the ?suprasylvian su lcus except for N. tenebrosus and S. gracilis , although this and previous studies (Gazin, 1965; Radinsky, 1970; Gurche, 1978) have shown that the preservation of the ?suprasylvian sulcus is fairly variable intraspecifically. Thus, while the ?suprasylvian s ulcus has not been observed on A. parisiensis , N. antiquus, T. homunculus, or R.. viejaensis , this may be an artifact of sample size. How ever, other primates in which ?suprasylvian sulci have been identified include the plesiadapiforms M. annectens and pos sibly P. cookei and Megadelphus ludeliusi hypothesis is thus intriguing. Cerebral Proportions and Sexual D imorphism Roughly, the proportions of the endocasts of N. ten ebrosus look more variable when compared to the seemingly more uniform proportions of the S. gracilis endocasts ( Figure 3 7). In general, the endocast width to length ratios seem to show the most differences among specimens of N. tenebrosus , with the ratio of AMNH 127167 (with or without olfactory bulbs) approaching the width to length ratio of the S . gracilis specimens . Meanwhile the female N. tenebrosus specimens, USNM V 23278 and USNM V 23277 having relatively more elongate endocasts than even A. parisie nsis , when the olfactory bulbs are not included.

PAGE 81

81 It is possible that some distortion in the braincase of the N. tenebrosus specimens could account for some of the variation in shape, particularly in AMNH 127167. The result that the height to width rati o of USNM V 23277 and USNM V 23278 are identical whereas the height to width ratio of AMNH 127167 is lower and suggests that AMNH 127167 is dorsoventrally flatter for its lateral width. This could signify that there may have been some dorsoventral co mpression during the fossilization of the specimen. On a similar token, the broken, turned in zygomatic arches of USNM V 23277 could be taken as a sign that perhaps this specimen was subject to some lateral compression during fossilization. However, if sig nificant lateral compression had occurred and led to distortion in the endocast of USNM V 23277, fractures in the brain case would be expected, of which there were none. Furthermore, the fractures present in the cranium of AMNH 127167 are minimal and conce ntrated in areas other than the brain case. In addition, it remains that the absolute height of USNM 23277 and USNM 23278, a specimen that seemingly has not been compressed laterally, are very similar . In fact, the height to length ratio of USNM 23278 is a ctually greater in USNM 23277, signifying that for its length, the height of USNM 23278 is greater than USNM 23277 despite the lack of lateral distortion that could increase the height of the endocast. With this evidence, it is arguable that the proportion al differences in the endocasts of N. tenebrosus may represent variation that is not accountable by simple post mortem distortion . Interestingly, the individual of S. gracilis with the lowest height to width ratio and the lowest height to length ratio is the male specimen USNM 17994, signifying that relative to the other S. gracilis specimens (the female USNM 179 96 and the juvenile UM 32773), it is the flattest endocast in the virtual sample. AMNH 127167, the male N. tenebrosus in the sample, also has the lowest height to width ratio and the lowest height to length ratio relative to the

PAGE 82

82 female specimens in the N. tenebrosus sample (USNM 23277 and USNM 23278). The cranium of USNM V 17994 is well preserved and does not show any signs of fracture in the region of the brain case, suggesting that dorsoventral crushing is not a factor in the proportions of this endocast. N. tenebrosus and S.gracilis endocasts in the sample may be related to the sexually dimorphic s kull shape. Alexander (1994) notes that male N. tenebrosus and S. gracilis crania appear to have broader, more flared cheeks relative to the females, which have less prominently forward facing cheeks. This difference is evident in our sample a s well (see F igure 3 2 ). Alexander (1994) also proposes that N. tenebrosus exhibits a greater degree of craniofacial sexual dimorphism than S. gracilis. The sexually dimorphic shape of the cranium may affect the overall width of the brain cavity, produc ing more pronoun ced intraspecific variation in the endocast shapes of the more sexually dimorphic N. tenebrosus , with and a lessened effect on the relatively less sexually dimorphic S. gracilis . Thus, the level of sexual dimorphism between the two species may explain the some of the differences in the level of intraspecific variation in cerebral proportions. Ontogeny and Endocast M orphology There are relatively few studies that examine ontogenetic changes in endocast or brain morphology (Macrini et al., 2007 ). Regardless , the juvenile S. gracilis endocast, UM 32773, is remarkably similar to the adult endocasts despite having the smallest absolute volume compared to USNM V 17994 and USNM V 17996 ( Table 3 3) and some evidence that the individual still had brain growth left if the patterns of growth in S. gracilis was comparable to the growth of Saimiri sciurus , the modern squirrel monkey ( Figure 3 1). While it is not entirely clear how much more growth and development the brain would have undergone until the individual matur ed into adulthood, it does seem likely that the specimen still had some b rain growth left

PAGE 83

83 when it die d , especially considering that UM 32773 has the smallest absolute brain volume of the S. gracilis specimens in the sample but the highest body mass predict ed by upper molar area. Morphologically, t he endocast of UM 32773 primarily differs from the adult specimens in minor proportions and the preservation of suture lines. The ratios of width, length, and height are not partic ularly different from the two adu lt S. gracilis . Casts of the coronal suture, sagittal suture, and parietal interparietal suture are clearly visible on the juvenile endocast. A ridge along the coronal suture is actually visible on all of the notharctine endocasts in the virtual sample exc ept for USNM V 17994, which is not surprising given the observation that the crania of S. gracilis and N. tenebrosus tend to separate at the coronal suture after death (Alexander, 1994) . This implies that the coronal suture is not obliterated in the adult cranium. No other suture is visible on the endocasts of the adult notharctine specimens, with the possible exception of the interparietal occipital suture, which happens to coincide with the location of the ?fissura prima in all of the notharctine endocast s. Gurche (1978) argues that the consistent occurrence of this indentation in the cerebellum in all known endocasts of S. gracilis suggests that this is a true fissure and not a dent caused by distortion of the occipital an d parietal bones during fossiliza tion. The evidence from the transparent endocasts and the universal presence of this feature in the notharctine endocasts in this study support this hypothesis. Interestingly, the natural endocast of Notharctus tenebrosus by Gurche (1978 and 1982) to appear as if casts of the coronal and sagittal suture are preserved, and the suture across which the partial endocast broke off from a the rest of the specimen appears to be around where the suture of the interparietal and parieta l bones would have been. While this specimen was not available for this study due to its mysterious origins and lack of information N.

PAGE 84

84 tenebrosus with unfused cranial sutu res. It is reasonable to hypothesize that some isolated endocasts may be found as a result of their separation from unfused crania l bones of juveniles. Stratigraphy and Endocast M orphology While this study presents the largest sample of consistently visual ized endocasts of N. tenebrosus and S. gracilis , it is unlikely that a sample size of n=3 for each species captures the full picture of variation of brain morphology for these taxa. In addition to possible variation from sex or ontogeny, variation in endoc ast morphology of these fossil taxa may also be affected by the stratigraphic level in which they were found. Notharctus tenebrosus have been found in localities in theBr2 biochron ( Gunnell; 2002; Gunnell et al., 2008), which may span some 1 million years (Robinson et al., 2004; see Figure 1 1 ). Likewise, Smilodectes gracilis has been identified in localities from the Br2 and Br 3 biochron, which expands the ra nge of the species to about 2 million years if the age. Gurche (1978) made the point that variatio n in the endocast morphology of fossil primates is not unexpected given the span of time in which a species existed. It is reasonable to expect significant variation among individuals from populations of disparate time periods, especially when comparable v ariation in brain morphology exists in relatively more concurrent populations (Gurche, 1978, Macrini 2007 ). While the locality where AMNH 127167 was collected is fairly precise (Alexander and Hamrick, 1994), the localities of other two N. tenebrosus specim ens are less so. However, two points of evidence support that all of the Notharctus tenebrosus specimens in this sample were mo st likely found in Bridger B/Br 2 biochron. Firstly, USNM V 23277 and USNM V 23278 were located 10 miles east of Lyman, WY, and 0. 5 miles South of Church Buttes, respectively, which are which are located within an area where the Bridger B/ Br2 biochron is exposed (Gunnell, 1998). Furthermore, the specimens were collected during an expedition to Wyoming and Utah by

PAGE 85

85 C. Lewis Gazin, cu rator of vertebrate paleontology at the U.S. National Museum, his wife Elisabeth, and Franklin Pearce, chief preparator of vertebrate paleontology, in July August 1959, with a primary focus on collecting from Bridger B localities (Lay; Dunkle, 1959). Pres umably, it was during this same expedition that USNM V 23276, the specimen of S. gracilis described in Gazin (1965) was found by Elisabeth Gazin in a Bridger B locality in or near section T 16 N, R 110 W, north of Cedar Mountain, which is south east of Lym an, WY. The S. gracilis specimens in the virtual sample appear to come from more varied strata. UM 32773 was collected in a University of Michigan locality in the Br 2 biochron, as was USNM 17996, found near Pinnacle Rock, located in the Bridger B. However , USNM 17994 was noted in the original collection tag to have been located in the Bridger C formation in a locality on Twin Butt es, which corresponds to the Br 3 biochron. While the skulls used in this study may thus be confined to rough stratigraphic laye rs, no more precise data exists. It is interesting that the S. gracilis endocast from the Br 3 layer, USNM V 17994, is more or less similar to the other S. gracilis endocasts under discussion, although it seems to be the only specimen reasonably separated in time. It may suggest that the endocrainal proportions of S. gracilis were relatively more stable through time than N. tenebrosus . Is S. gracilis More E ncephalized than N. tenebrosus ? Earlier comparative studies of notharctine endocasts suggested that th e EQ of N. tenebrosus was slightly higher than the EQ of S. gracilis (Gurche, 1978; Gurche, 1982; Martin, 1990), whereas the results of this study suggest that when body masses estimated from dental proxies are used, N. tenebrosus has a definitively lower EQ than S. gracilis . Although previous results were probably affected by the suspected overestimation in endocranial volume of N. tenebrosus estimation methods, discrepancy in b ody mass results has a high degree of influence in the EQ

PAGE 86

86 differences observed between the notharctines in this study. However imprecisely, previous workers have noted the similarity in body size of N. tenebrosus and S. gracilis (Jerison, 1973; Rad insky, 1 977; Gurche, 1978) . It is pertinent to ask whether the body masses of these two notharctines were indeed different enough to warrant different EQ . The results of this study suggest that dental proxies overestimate the body mass of N. tenebrosus relative to S. gracilis , especially considering that the body mass estimate s calculated using postcranial variables suggest that N. tenebrosus and S. gracilis have overlapping body mass estimates when their dental estimates are consistently non overlapping. If dent ally derived body mass estimates of N. tenebrosus are indeed unreliable, it then becomes necessary to consider which postcranial variables should be used as proxies of body mass instead. The results of this study suggest that several of the postcraninal e quations are probably more reliable than others. Bone diamaters are probably more reliable than bone lengths in determining body mass estimates for primates, as their locomotive specializations may result in differing limb lengths that may affect body mass estimation. In particular, N. tenebrosus and S. gracilis are noted to have longer hind limbs than forelimbs, which probably reflect their leaping specializations (Covert, 1986). For that same reason, the all primate equation for body mass is probably more reliable than the all mammal equations for body mass, as other typical mammals may not exhibit the primate like patterns of limb proportions. Not surprisingly, some of the highest body postcranial body mass estimates that have resulted were calculated fro m the all mammal femur length regression equation of Gingerich (1990), probably reflecting that these two notharctines had longer hindlimbs than expected for a mammal of their body size. When the primates only equations of Boyer (2009) are used, femur leng th gives much lower body mass estimates; but another lower limb measurement, the tibia diameter, gives a much higher body mass estimate for

PAGE 87

87 S. gracilis than N. tenebrosus , an unexpected result. The tibial diameter of S. gracilis may have been affected by l arger anterior crest on the tibia relative to N. tenebrosus, and thus is probably not a reliable measure. Despite having a small range of body mass resuts, ulnal diameter is probably not a good proxy as well, for most of the weight of the lower forelimbs a re borne on the radius, not the ulna. While humerus diameter would seem like a good candidate for an appropriate body mass proxy, the range of body masses obtained are rather disparate, suggesting that this proxy may also be less reliable. If one variable of those explored in this study must be chosen as a proxy for body mass, femoral diameter may be ideal measurement. Of all the equations presented in Gingerich (1990), femoral diameter has the highest r 2 value at .991. Body mass estimates of S. gracilis a nd N. tenebrosus calculated using femoral diameter are distributed rather closely together in both the all mammal and primates only equations, and appear to be the least fraught with problems of unusual crests affecting diameters. If the primates only regr ession results for femoral diameter N. tenebrosus and S. gracilis , this would suggest that their body masses overlap each other. Of course, a more intensive study of larger samples of associated notharctine skeletons is necessary to statistically support the conclusions presented here, but this preliminary analysis at least seems to suggest that N. tenebrosus and S. gracilis had overlapping body masses, which would minimize differences in the encephalization quotients of these two species. However, i f the relative difference in N. tenebrosus and S. gracilis is real, this would be a surprising result considering their purported phylogenetic closeness. Disparate encephalization quotients in these species may s uggest some ecological or behavioral differences that may have affected their brain size. Typically, differences in diet, reliance on vision, and social structure are

PAGE 88

88 hypothesized to affect the level of encephalization in primate taxa (Silcox et al., 2010) . It has been noted that among small bodied mammals and primates, folivores are relatively less encephalized in comparison to closely related non folivorous taxa (Harvey et al . , 1980). Based on dental characteristics and analyses, N. tenebrosus and S. grac ilis are both believed to have been folivorous (Covert, 1986 and 1995; Gilbert, 2005; Sauther and Cuozzo, 2012; Fleagle, 2013), making dietary difference an unlikely candidate for encephalization differences without taking other factors into account. The evolution of vision systems in primates has also been linked to encephalization that the evolution of more complex or acute vision systems would coincide wit h enlargement of brain areas that process vision (Barton, 2004; Kirk, 2006a). Incidentally, the primates with the highest visual acuity (extant haplorhines, which are almost all diurnal) also have the largest encephalization quotients (Kay and Kirk, 2000; Kirk and Kay, 2004). Therefore, it has been hypothesized that there has been an evolutionary selection for the enlargement of vision processing areas that could have influenced encephalization of the rest of the brain (Kirk, 2006a). Notharctus tenebrosus a nd S. gracilis are thought to have both been diurnal (Covert, 1986, 1995, 2004; Fleagle, 2013), although it is unlikely that the two species were highly specialized for visual acuity based on the results of this study. Lastly, the social brain hypothesis posits that primate brains expanded along with the evolution of intricate social systems as measured by group size (Dunbar and Shultz, 2007). Amongst modern primates, there is a positive relationship between relative brain size and sociality (Dunbar and Sh ultz, 2007), and increase in brain size of primates may have been driven by some facet of it (Shultz and Dunbar, 2010). Unfortunately, social behavior is one of the more

PAGE 89

89 difficult aspects of an extinct animal to discern, but differential canine size within a species is an indicator of sexual dimorphism, which has been used to infer some degree of male male competition and the possibility of polygyny (Plavcan, 2000, 2004). While speculating on the behavior of fossil taxa based on dimorphism is problematic, H arvey et al . (1980) found that primates that form pair bonds may have a relatively smaller brain size although it was noted that the sample set was rather small, perhaps making this association dubious. Indeed, more recent studies (Schillaci, 2006; Schilla ci, 2008) have revealed that monogamous primates have relatively larger brains and neocortices than non monogamous species that practice extensive male male competition (particularly among hominoids), which may seem at first to contradict the social brain hypothesis without considering whether monogamous or non monogamous mating systems require more complex behaviors that would select for increased encephalization. Both N. tenebrosus and S. gracilis exhibit secondary sexual dimorphism, although N. tenebrosu s seems to express more canine and craniofacial sexual dimorphism than S. gracilis (Alexander, 1994). While more monogamous anthropoids tend to exhibit lesser degree of sexual dimorphism (Plavcan, 2002; Fleagle, 2013), it is difficult to say whether differ ences in the level of dimorphism between these two notharctines could have played any role in possible encephalization differences. It is apparent that the two notharctines were rather similar in many of the traits that may be used to assess possible encep halization differences. Overall, regardless of which body mass alth N. tenebrosus and S. gracilis do not approach those of extant haplorhines. This is not surprising, considering notharctines, as adapiforms, are considered stem

PAGE 90

90 strepsirrhines. While this study suggests that the notharctines were not much m ore encephalized than plesiadapiforms, there is some evidence that the notharctine brains were becoming more visually oriented based on the expansion of the temporal and occipital pole and reduced olfactory bulb size to brain mass and body mass relative t o plesiadapiforms.

PAGE 91

91 CHAPTER 5 CONCLUSIONS This study documents the first virtual endocasts of the notharctine primates Notharctus tenebrosus and Smilodectes gracilis , providing a larger sample size of consistently reconstructed endocasts that may be more comparably quantified for comparisons. Relative to previous works, volumes could be more accurately measured, endocast surface morphology could be visualized at a higher resolution, and the sample size of specimens available for study was expanded due to the non destructive technique of endocast extraction. This work confirmed that the endocranial morphology of N. tenebrosus was consistent with those of other adapiforms , such as Smilodectes gracilis and the adapine adapoid Adapis parisiensis , in having sm all frontal lobes relative to extant primates, non overlap of the cerebrum over the olfactory bulbs or cerebellum, and a relatively smaller olfactory bulbs and more expanded temporal and occipital lobes relative to plesiadapiforms. While natural endocasts of Smilodectes gracilis have been known and discussed in the literature for some 50 years, this study revealed that additional specimens do not particularly contradict the findings of other workers. However, USNM V 17994 may preserve the first hint of a ?S ylvian sulcus, which has never been observed in this species before. The evidence that ?vessels seem to be located on and near the area that the Sylvian sulcus would be expected in S. gracilis and N. tenebrosus, as well as the evidence that the Sylvian sul cus appears more like a fossa and is weaker in N. tenebrosus compared to Adapis parisiensis and other euprimate endocasts, could suggest that the Sylvian sulcus was partially (in the case of N. tenebrosus ) or completely (in the case of S. gracilis ) obscure d by sub dural blood vessels, as originally suggested by Radinsky (1977). However, the appearance of these ?vessel over the frontal temporal region of the endocast, as well as the

PAGE 92

92 ?suprasylvian sulcus, which Radinsky (1977) to some extent associated with t he lack of the Sylvian sulcus in S. gracilis is variable among the notharctine endocasts in this sample. Interestingly, the endocasts of S. gracilis are more similar to one another in general shape and proportions compared to the endocasts of N. tenebrosu s. The virtual endocast of AMNH 127167 suggests a Smilodectes like dorsal outline, while USNM V 23277 has a more Adapis like interesting because the three N. tenebrosus skull s are mo st likely constrained to the Br 2 biochron or the Bridger B layer stratigraphically, while the S. gracilis spe cimen may come from both the Br 2 a nd the chronologically later Br 3 (Bridger C). In addition, all N. tenebrosus specimens in the sample were adults; despite the fact that UM 32773 represents a juvenile specimen with still erupting teeth and unfused cranial sutures, the surface morphology and proportions are very similar to adult S. gracilis endocast. This could suggest that the brains of N. te nebrosus were more variable in surface morphology, volume, and proportions relative to S. gracilis . It is possible that it may be affected by the more pronounced sexual dimorphism in cranial shape of Notharctus tenebrosus relative to S. gracilis . It is d ifficult to assess the relative encephalization of these notharctines relative to each other and to other primates without an accurate estimation of their body masses. While the only proxy that all specimens in the sample possessed were dental, this study suggests that estimates of body mass derived from dental proxies may be overestimating the body mass of Notharctus tenebrosus . Perhaps, then, dental proxies are useful for determining relative size of the animal intraspecificially, but may not be as reliab le for interspecific variation. Overall, findings suggest that N. tenebrosus was less encephalized than S. gracilis, if dental proxies may be trusted . Postcranial estimations of body mass suggest that N. tenebrosus and S. gracilis had overlapping

PAGE 93

93 body mass es, but this could still imply that the encephalization quotient of N. tenebrosus was lower than S. gracilis difference in encephalization is true, it would be an intriguing finding as these species are purported to be closely related and would be expected to have similar encephalization quotients. Similarities in proxies for diet, level of visual acuity, and social systems seem to suggest that these species shared similar habits, alt hough there are many unknowns that cloud the true picture of differences, if any, that may have played a role in the evolution of their respective encephalizations. Overall, however, it is likely that the brains of the notharctines N. tenebrosus and S. gr acilis were less encephalized than any extant primate, and their relative size and morphologies were still primitive and quite similar to plesiadapiforms even if there does appear to be some evidence that the visual centers of the brain was expanding with a lessened emphasis in olfaction for these early euprimates .

PAGE 94

94 APPENDIX A TABLES Table A 1 . Canine dimensions of Notharctus tenebrosus and Smilodectes gracilis Tooth Dimension n Range SD V N. tenebrosus male C 1 L 5 4.0 5.0 4.7 0.40 8.5 C 1 W 5 3.0 3.5 3.2 0.07 2.2 C 1 H 5 5.5 8.5 7.0 1.10 15.7 C 1 L 6 3.0 4.5 3.8 0.46 12.1 C 1 W 5 3.0 3.0 3.0 0.00 0.00 C 1 H 4 7.0 8.5 7.8 0.56 7.2 N. tenebrosus female C 1 L 1 3.0 C 1 W 1 2.5 C 1 H 1 5.0 C 1 L 3 2.5 3.0 2.7 0.24 8.9 C 1 W 3 2.0 2.5 2.2 0.24 10.9 C 1 H 2 4.5 5.0 4.8 S. gracilis male C 1 L 1 4.0 C 1 W 1 2.5 C 1 H 1 S. gracilis female C 1 L 1 3.0 C 1 W 1 2.0 C 1 H 1 5.0 C 1 L 1 3.0 C 1 W 1 2.0 C 1 H 1 4.0 All measurements are in millimeters (mm). C 1 = maxillary canine; C 1 = mandibular canine; L=length; W=width; H=height; n=sample size; = sample mean; SD= standard deviation; V= coefficient of variation. All data from Alexander (1994).

PAGE 95

95 A 2. Previous body mass, endocranial volume, encephalization quotient, and calculation method for fossil primates. Taxon Source Specimen BM Method 1 E CV Method 2 EEQ OBV N. tenebrosus 3 2500 CS 10.43 WD 0.58 236.60 N. tenebrosus 4 1990 CPM 10.40 F3 0.68 Smilodectes 1 2540 CSN 9.50 DGI 0.52 Smilodectes 2 1700 CSN 9.60 DGI 0.67 S. gracilis 3 YPM 12152 2500 CSN 9.95 WD 0.55 S. gracilis 3 USNM 23276 2400 CSN 9.12 WD 0.52 S. gracilis 4 1960 CPM 9.50 F3 0.63 Adapis 1 2540 CSN 9.00 DGI 0.49 Adapis 2 1700 CSN 9.00 DGI 0.67 A. parisiensis 3 BM M 20192 2500 CSN 8.31 WD 0.46 266.60 A. parisiensis 4 2350 CPM 8.80 F5 0.51 A. parisiensis 5 BMNH M1345 2450 CPM2 8.80 MSV 0.50 A. parisiensis 5 M 538 2350 CPM2 8.80 MSV 0.51 Rooneyia viejaensis 6, 10 TMM 40688 7 381 CPGLS 7.23 DV 1.62 68.00 Rooneyia viejaensis 6,10 TMM 40688 7 1821 GM 1 7.23 DV 0.51 68.00 Ignac ius graybullianus 7 USNM 421608 375 GM 1 2.41 DV 0.48 181.33 Microsyops annectens 8 UW 12362 2568 GM 1 5.90 DV 0.32 300.55 Plesiadapis tricuspidens 9 MNHM CR 125 2935 GM 1 5.21 DV 0.26 136.00 1= Radinsky, 1977; 2= Jerison, 1979; 3= Gurche, 1982; 4= Martin, 1990; 5= Gingerich and Martin, 1981 ; 6= Silcox et al., 2009a; 7= Silcox et al., 2009b; 8= Silcox et al., 2010; 9= Orliac et al., 2010; 10= Kirk et al., 2014; BM= body mass estimate in grams (g); CPGLS= estimated from cranial length using a combiled phylogenetic least squares regression from Silcox et al., 2009a; CPM= took the mean of body masses estimated from cranial proxies of body mass from Martin (1990); CPM 2= took the mean of body masses estimated from cranial proxies of body mass from Gingeri ch and Martin ( 1981 ); CS= compared to body mass of a similar sized modern primate analog using size of reconstructed skeleton; CSN= compared the reconstructed skeleton of Notharctus to a similar sized modern primate analog and assumed that the modern prima te, Notharctus , and the specimen in question were approximately of the same body mass; DGI= double graphic integration; DV= volume extracted from a digital endocast; ECV=endocranial (endocast) volume in cubic centimeters (cm 3 tion quotient; GM 1 = estimated using body mass equation using M 1 area from Gingerich et al. (1982); F3= from source 3 (Gurche, 1982); F5= from source 5 (Gingerich and Martin, 1981 ); Method 1= method in which BM was estimate; Method 2= method in which ECV a nd OBV was estimated; OBV= olfactory bulb volume in cubic millimeters (mm 3 ); MSV= volume estimated from filling the cranial cavity with mustard seeds; SS= judges to be slightly smaller in body mass than YPM 12152; WD= water displacement of reconstructed en docast .

PAGE 96

96 Table A 3. Average body mass, endocranial volume, encephalization quotient of extant primate species. Taxon Category Source BM ECV EEQ n Allenopithecus nigroviridis H 1 4655 59.07 2.07 3 Allenopithecus nigroviridis H 2 4655 58.02 2.04 13 Alou atta belzebul H 2 6395 52.84 1.47 11 Alouatta caraya H 2 5383 52.63 1.66 9 Alouatta guariba H 2 5175 51.70 1.68 7 Alouatta palliata H 1 6250 56.05 1.58 6 Alouatta palliata H 2 6250 49.88 1.41 31 Alouatta pigra H 2 8915 51.13 1.11 8 Alouatta seniculus H 1 5950 54.84 1.61 6 Alouatta seniculus H 2 5950 55.22 1.62 37 Aotus azarae H 2 1205 20.67 1.97 6 Aotus azarai H 1 1205 20.67 1.97 6 Aotus lemurinus H 2 734 16.30 2.24 38 Aotus trivirgatus H 1 775 17.65 2.34 6 Aotus trivirgatus H 2 989 16.85 1.86 1 2 Arctocebus aureus S 2 210 5.88 2.04 5 Arctocebus calabarensis S 1 309 7.02 1.83 6 Arctocebus calabarensis S 2 309 6.92 1.81 21 Ateles belzebuth H 2 8167 117.02 2.71 10 Ateles fusciceps H 1 9025 116.90 2.51 6 Ateles fusciceps H 2 9025 114.24 2.46 9 Ateles geoffroyi H 1 7535 109.20 2.68 7 Ateles geoffroyi H 2 7535 105.09 2.58 20 Ateles hybridus H 2 7862 103.05 2.45 8 Ateles paniscus H 2 8280 103.85 2.38 23 Avahi laniger S 1 1175 9.75 0.95 6 Avahi laniger S 2 1207 9.86 0.94 14 Brachyteles arachn oides H 1 8840 119.40 2.61 2 Brachyteles sp. H 2 8840 116.72 2.55 6 Bunopithecus hoolock H 2 6728 110.68 2.96 8 Cacajao calvus H 2 3165 76.00 3.55 15 Cacajao melanocephalus H 1 2935 68.73 3.39 8 Cacajao melanocephalus H 2 2935 68.77 3.40 9 Callicebus caligatus H 2 880 17.25 2.08 4 Callicebus discolor H 2 875 17.53 2.12 5 Callicebus lugens H 2 1090 18.97 1.95 14 Callicebus medemi H 2 1157 19.44 1.91 7 Callicebus torquatus H 1 1245 19.49 1.82 8 Callimico goeldii H 1 484 11.57 2.17 7 Callimico goel dii H 2 484 11.43 2.14 10 Callithrix argentata H 1 345 8.14 1.96 9 Callithrix argentata H 2 345 7.95 1.91 11

PAGE 97

97 Table A 3. Continued Taxon Category Source BM ECV EEQ n Callithrix jacchus H 1 321 7.77 1.97 7 Callithrix jacchus H 2 320 7.24 1.84 7 Callith rix penicillata H 2 328 7.32 1.83 32 Callithrix pygmaea H 2 116 4.17 2.25 11 Cebuella pygmaea H 1 116 4.37 2.36 6 Cebus albifrons H 1 2735 63.36 3.30 6 Cebus albifrons H 2 2735 65.45 3.41 32 Cebus apella H 2 2936 66.63 3.29 48 Cebus capucinus H 1 311 0 75.86 3.59 6 Cebus capucinus H 2 2861 72.93 3.67 28 Cebus libidinosus H 2 2510 64.69 3.59 47 Cebus nigritus H 2 2731 67.07 3.49 87 Cebus olivaceus H 2 2931 69.84 3.45 14 Cebus xanthosternos H 2 2440 66.09 3.74 11 Cercocebus atys H 2 8600 94.68 2.11 20 Cercocebus galeritus H 1 7435 114.10 2.83 6 Cercocebus torquatus H 1 7485 107.20 2.65 6 Cercocebus torquatus H 2 7485 105.99 2.62 7 Cercopithecus albogularis H 2 5620 70.10 2.14 33 Cercopithecus ascanius H 2 3714 59.58 2.47 23 Cercopithecus campb elli H 2 3600 57.39 2.44 33 Cercopithecus cephus H 2 3585 65.26 2.78 26 Cercopithecus diana H 2 4550 62.61 2.24 20 Cercopithecus lhoesti H 2 4710 74.20 2.58 17 Cercopithecus lowei H 2 3187 55.64 2.59 15 Cercopithecus mitis H 2 6109 71.33 2.05 18 Cerc opithecus mona H 2 3719 61.84 2.56 6 Cercopithecus neglectus H 2 5450 65.97 2.06 29 Cercopithecus nictitans H 2 5465 71.13 2.22 20 Cercopithecus petaurista H 2 3609 55.08 2.33 39 Cercopithecus pogonias H 2 3580 59.56 2.54 12 Cercopithecus wolfi H 2 33 90 61.45 2.73 20 Cheirogaleus major S 1 400 5.83 1.26 6 Cheirogaleus major S 2 400 5.81 1.25 14 Cheirogaleus medius S 1 283 3.24 0.90 6 Cheirogaleus medius S 2 140 2.60 1.22 7 Chiropotes chiropotes H 2 2740 57.04 2.96 6 Chiropotes israelita H 2 2274 55.13 3.29 16 Chiropotes satanas H 1 2740 56.79 2.95 6 Chlorocebus aethiops H 1 3620 69.77 2.95 6 Chlorocebus aethiops H 2 3720 65.00 2.69 53 Chlorocebus pygerythrus H 2 4324 62.58 2.32 61

PAGE 98

98 Table A 3. Continued Taxon Category Source BM ECV EEQ n Chl orocebus sabaeus H 2 4312 64.91 2.41 7 Colobus angolensis H 2 8625 77.70 1.73 46 Colobus guereza H 1 8895 82.95 1.80 6 Colobus guereza H 2 8589 74.39 1.66 37 Colobus polykomos H 1 9100 78.86 1.69 7 Colobus polykomos H 2 9100 73.83 1.58 32 Colobus sat anas H 2 8910 74.90 1.63 13 Colobus vellerosus H 2 7820 73.07 1.75 10 Daubentonia madagascariensis S 1 2555 44.87 2.46 7 Daubentonia madagascariensis S 2 2555 44.85 2.45 10 Erythrocebus patas H 1 9450 94.38 1.96 6 Erythrocebus patas H 2 9450 97.73 2.0 3 23 Eulemur coronatus S 1 1655 20.31 1.53 6 Eulemur coronatus S 2 1180 20.65 2.00 7 Eulemur fulvus S 1 2215 24.75 1.51 9 Eulemur fulvus S 2 2292 25.77 1.53 8 Eulemur macaco S 1 2440 24.80 1.40 7 Eulemur macaco S 2 2390 24.51 1.41 8 Eulemur mongoz S 1 1620 20.82 1.60 8 Eulemur mongoz S 2 1212 20.17 1.92 14 Eulemur rubriventer S 1 1960 26.33 1.75 7 Eulemur rubriventer S 2 1960 26.23 1.75 8 Eulemur rufus S 2 2220 25.40 1.54 4 Euoticus elegantulus S 1 274 5.85 1.67 6 Euoticus elegantulus S 2 274 5 .53 1.58 14 Euoticus pallidus S 2 300 5.19 1.38 9 Galago alleni S 1 273 5.85 1.68 6 Galago cameronensis S 2 252 5.58 1.70 8 Galago demidoff S 2 75 2.65 1.98 25 Galago gabonensis S 2 273 5.36 1.53 7 Galago moholi S 1 180 3.83 1.49 10 Galago moholi S 2 148 3.71 1.67 61 Galago senegalensis S 1 213 4.65 1.60 6 Galago senegalensis S 2 194 3.96 1.46 199 Galago zanzibaricus S 2 143 3.51 1.62 6 Galagoides demidoff S 1 62 2.72 2.33 6 Galagoides zanzibaricus S 1 143 3.52 1.63 6 Gorilla beringei H 2 13000 0 491.27 1.47 6 Gorilla gorilla H 1 121000 491.30 1.55 6 Gorilla gorilla H 2 120950 490.41 1.55 56 Hapalemur griseus S 1 945 13.77 1.57 10 Hapalemur griseus S 2 709 14.09 1.99 15

PAGE 99

99 Table A 3. Continued Taxon Category Source BM ECV EEQ n Hylobates agi lis H 2 5850 91.16 2.70 16 Hylobates klossii H 1 5795 88.84 2.65 6 Hylobates klossii H 2 5795 87.99 2.63 6 Hylobates lar H 1 5620 101.50 3.10 6 Hylobates lar H 2 5595 101.87 3.12 207 Hylobates muelleri H 2 5821 85.13 2.53 11 Hylobates pileatus H 2 54 70 84.69 2.64 9 Hylobates syndactylus H 1 11300 128.40 2.34 6 Indri indri S 1 6335 36.58 1.02 6 Indri indri S 2 6335 34.81 0.97 16 Lagothrix cana H 2 8215 99.34 2.29 15 Lagothrix lagotricha H 1 7150 100.90 2.58 6 Lagothrix lagotricha H 2 7150 96.50 2 .47 9 Lemur catta S 1 2210 23.79 1.45 7 Lemur catta S 2 2210 22.90 1.40 10 Leontopithecus chrysomelas H 1 578 11.98 1.97 3 Leontopithecus rosalia H 1 609 13.04 2.06 9 Leontopithecus rosalia H 2 609 12.83 2.03 12 Lepilemur leucopus S 1 606 6.85 1.09 7 Lepilemur leucopus S 2 606 6.87 1.09 9 Lepilemur mustelinus S 1 777 9.56 1.26 5 Lepilemur mustelinus S 2 777 9.56 1.26 5 Lophocebus albigena H 1 7135 95.71 2.45 6 Lophocebus albigena H 2 6590 93.97 2.55 21 Lophocebus aterrimus H 1 6800 104.70 2.78 5 Lophocebus aterrimus H 2 6800 101.59 2.69 6 Loris lydekkerianus S 2 267 6.34 1.85 11 Loris tardigradus S 1 193 5.69 2.11 6 Loris tardigradus S 2 193 5.87 2.18 5 Macaca arctoides H 2 10300 100.70 1.96 34 Macaca fascicularis H 1 4475 68.70 2.48 6 Mac aca fascicularis H 2 4251 63.98 2.40 97 Macaca fuscata H 2 9515 102.92 2.13 22 Macaca leonina H 2 5642 85.60 2.61 7 Macaca mulatta H 2 6793 88.98 2.36 103 Macaca nemestrina H 1 8850 109.10 2.38 6 Macaca nemestrina H 2 8821 105.59 2.31 22 Macaca nigra H 2 7680 94.90 2.30 3 Macaca radiata H 2 5084 74.87 2.46 9 Macaca sylvanus H 2 12078 93.20 1.62 5 Mandrillus sphinx H 1 22250 162.10 1.79 6 Mandrillus sphinx H 2 23600 153.88 1.63 24

PAGE 100

100 Table A 3. Continued Taxon Category Source BM ECV EEQ n Microceb us murinus S 1 61 1.66 1.44 7 Microcebus murinus S 2 120 1.63 0.86 17 Microcebus rufus S 1 43 1.63 1.83 5 Microcebus rufus S 2 43 1.72 1.95 9 Miopithecus ogouensis H 2 1750 37.35 2.70 22 Miopithecus talapoin H 1 1250 41.34 3.84 9 Mirza coquereli S 1 315 5.81 1.50 6 Mirza coquereli S 2 168 5.80 2.38 8 Nasalis larvatus H 1 15110 96.05 1.41 6 Nasalis larvatus H 2 14561 92.30 1.39 45 Nomascus gabriellae H 2 7365 119.38 2.98 6 Nycticebus bengalensis S 2 1060 13.49 1.42 4 Nycticebus coucang S 1 653 10 .27 1.54 8 Nycticebus coucang S 2 653 10.13 1.52 30 Nycticebus pygmaeus S 1 307 7.23 1.90 6 Nycticebus pygmaeus S 2 307 7.23 1.90 6 Otolemur crassicaudatus S 1 1150 10.82 1.07 13 Otolemur crassicaudatus S 2 1150 11.78 1.16 15 Otolemur garnettii S 1 7 64 10.87 1.45 7 Otolemur garnettii S 2 764 11.50 1.54 29 Pan paniscus H 2 39100 341.29 2.48 10 Pan troglodytes H 1 38200 400.90 2.96 6 Pan troglodytes H 2 44967 368.35 2.41 115 Papio anubis H 1 19200 171.10 2.10 6 Papio anubis H 2 18150 167.42 2.15 3 7 Papio cynocephalus H 1 17050 186.60 2.51 6 Papio cynocephalus H 2 17150 163.19 2.18 24 Papio hamadryas H 2 14150 146.17 2.25 14 Papio ursinus H 2 22300 178.00 1.96 34 Perodicticus potto S 1 1230 12.68 1.19 11 Perodicticus potto S 2 835 12.42 1.55 7 0 Phaner furcifer S 1 460 6.47 1.26 5 Phaner pallescens S 2 339 6.68 1.63 7 Piliocolobus badius H 2 8285 63.59 1.46 40 Piliocolobus kirkii H 2 5630 57.25 1.75 13 Pithecia monachus H 2 2360 36.45 2.12 16 Pithecia pithecia H 1 1760 34.97 2.52 6 Pithec ia pithecia H 2 1760 32.26 2.33 18 Pongo abelii H 2 62815 389.50 1.99 25 Pongo pygmaeus H 1 57150 384.10 2.11 6 Pongo pygmaeus H 2 58542 377.38 2.03 81 Presbytis comata H 1 6700 72.79 1.95 6

PAGE 101

101 Table A 3. Continued Taxon Category Source BM ECV EEQ n P resbytis femoralis H 2 6848 73.74 1.95 14 Presbytis frontata H 1 5620 81.02 2.47 6 Presbytis frontata H 2 5615 81.02 2.48 6 Presbytis melalophos H 1 6530 66.22 1.81 6 Presbytis melalophos H 2 6560 64.85 1.77 21 Presbytis rubicunda H 2 6266 71.02 2.00 50 Presbytis siamensis H 2 5876 60.94 1.80 13 Presbytis thomasi H 2 6730 72.93 1.95 7 Procolobus badius H 1 8285 70.69 1.62 6 Procolobus verus H 2 4450 52.60 1.91 29 Propithecus deckenii S 2 3532 30.15 1.30 7 Propithecus diadema S 1 6100 40.06 1.15 6 Propithecus diadema S 2 6130 39.80 1.14 4 Propithecus edwardsi S 2 5682 39.49 1.20 4 Propithecus verreauxi S 1 3990 26.80 1.05 9 Propithecus verreauxi S 2 2955 26.21 1.29 7 Pygathrix nemaeus H 1 9720 88.20 1.80 8 Pygathrix roxellana H 1 14750 114.40 1.71 6 Pygathrix sp. H 2 9720 85.86 1.75 10 Saguinus fuscicollis H 1 351 7.83 1.86 6 Saguinus fuscicollis H 2 401 7.94 1.71 31 Saguinus geoffroyi H 2 517 10.14 1.81 24 Saguinus leucopus H 1 492 9.33 1.73 7 Saguinus leucopus H 2 525 9.70 1.71 12 Sag uinus midas H 1 545 9.65 1.66 6 Saguinus midas H 2 563 9.78 1.64 15 Saguinus mystax H 2 584 11.09 1.81 8 Saguinus niger H 2 375 9.48 2.15 7 Saguinus oedipus H 2 431 9.76 1.99 50 Saimiri oerstedii H 1 789 25.96 3.39 10 Saimiri oerstedii H 2 789 25.07 3.27 15 Saimiri sciureus H 1 721 27.50 3.84 6 Saimiri sciureus H 2 799 24.14 3.12 88 Semnopithecus dussumieri H 2 10664 90.09 1.71 10 Semnopithecus entellus H 1 11450 103.70 1.87 6 Semnopithecus priam H 2 8562 83.07 1.86 12 Simias concolor H 1 7975 5 9.09 1.39 6 Simias concolor H 2 7975 58.42 1.38 6 Symphalangus syndactylus H 2 11295 123.50 2.25 38 Tarsius bancanus H 1 123 3.24 1.67 6 Tarsius bancanus H 2 126 3.16 1.61 19 Tarsius spectrum H 1 117 3.00 1.61 6

PAGE 102

102 Table A 3. Continued Taxon Category Source BM ECV EEQ n Tarsius syrichta H 1 126 3.46 1.76 6 Tarsius syrichta H 2 126 3.36 1.71 9 Tarsius tarsier H 2 117 3.06 1.64 5 Theropithecus gelada H 1 15350 131.80 1.91 8 Theropithecus gelada H 2 15350 133.33 1.94 27 Trachypithecus cristatus H 1 6185 64.88 1.85 6 Trachypithecus cristatus H 2 6394 57.86 1.61 30 Trachypithecus obscurus H 2 7056 62.12 1.60 29 Trachypithecus phayrei H 2 7475 72.84 1.80 9 Trachypithecus vetulus H 2 6237 61.29 1.73 6 Varecia variegata S 1 3490 32.25 1.40 7 Varecia variegata S 2 3575 32.12 1.37 9 1= Kirk, 2006b; 2= Isler et al., 2009; BM= Average body mass in grams (g); ECV= average endocranial volume in cubic centimeters (cm 3 s encephalization quotient; n= number of specimens in the sample.

PAGE 103

103 Tab le A 4 . Body mass, endocranial volume, and olfactory bulb volume of primates and Taxon Cat. BM ECV OBV ln ECV ln OBV ln BM Aethechinus algirus BI 700 3174.00 297.00 8.06 5.69 6.55 Crocidura occident. BI 28 408.00 37.80 6.01 3.63 3.33 C rocidura russula BI 11 178.00 16.40 5.18 2.80 2.40 Echinops telfairi BI 88 569.00 65.70 6.34 4.19 4.47 Erinaceus europaeus BI 860 2969.00 350.00 8.00 5.86 6.76 Hemicentetes semisp. BI 110 757.00 93.80 6.63 4.54 4.70 Setifer setosus BI 243 1404.00 210.0 0 7.25 5.35 5.49 Sorex araneus BI 10 188.00 14.20 5.24 2.65 2.33 Sorex minutus BI 5 103.00 8.70 4.63 2.16 1.67 Suncus murinus BI 35 354.00 34.40 5.87 3.54 3.56 Tenrec ecaudatus BI 832 2336.00 293.00 7.76 5.68 6.72 Alouatta seniculus H 6400 47749.00 45 .20 10.77 3.81 8.76 Aotus trivirgatus H 850 15229.00 60.30 9.63 4.10 6.75 Ateles geoffroyi H 8000 101034.00 92.60 11.52 4.53 8.99 Callicebus moloch H 650 14434.00 16.80 9.58 2.82 6.48 Callithrix jacchus H 260 7248.00 28.40 8.89 3.35 5.56 Cebus albifro ns H 3000 75592.00 37.20 11.23 3.62 8.01 Cebus sp. H 2600 68672.00 49.60 11.14 3.90 7.86 Cercopithecus ascanius H 3600 61610.00 96.50 11.03 4.57 8.19 Cercopithecus talapoin H 1000 36830.00 22.30 10.51 3.10 6.91 Cereocebus albigena H 7900 97603.00 121.0 0 11.49 4.80 8.97 Cereocebus albigena H 7900 97603.00 121.00 11.49 4.80 8.97 Cereopithecus mitis H 6500 71505.00 118.00 11.18 4.77 8.78 Colobus badius H 7000 73818.00 51.30 11.21 3.94 8.85 Gorilla gorilla H 125000 437433.00 294.00 12.99 5.68 11.74 Hom o sapiens H 65000 1251847.00 114.00 14.04 4.74 11.08 Lagothrix lagotricha H 5200 94939.00 83.00 11.46 4.42 8.56 Macaca mulatta H 6000 87896.00 84.30 11.38 4.43 8.70 Pan troglodytes H 46000 396255.00 267.00 12.89 5.59 10.74 Pithecia monacha H 1500 32836 .00 38.20 10.40 3.64 7.31 Saguinus oedipus H 405 10576.00 24.60 9.27 3.20 6.00 Saguinus tamarin H 340 9459.00 16.80 9.15 2.82 5.83 Saimiri sciureus H 680 20691.00 25.40 9.94 3.23 6.52 Tarsius syrichta H 88 3416.00 18.10 8.14 2.90 4.47 Chorotalpa stuhi manni PI 40 693.00 60.80 6.54 4.11 3.68 Desmana moschata PI 440 3620.00 142.00 8.19 4.96 6.09 Elephantulus fuscipes PI 57 1233.00 63.90 7.12 4.16 4.04 Galemys pyrenaicos PI 58 1230.00 39.20 7.11 3.67 4.05 Limnogale mergulus PI 92 1046.00 43.20 6.95 3.7 7 4.52 Neomys fodiens PI 15 299.00 15.90 5.70 2.77 2.72 Nesogale talazaci PI 50 741.00 74.60 6.61 4.31 3.92 Potamogale velox PI 660 3822.00 87.00 8.25 4.47 6.49

PAGE 104

104 Table A 4 . Continued Taxon Cat. BM ECV OBV ln ECV ln OBV ln BM Rhynchocyon stuhimanni PI 490 5680.00 427.00 8.64 6.06 6.19 Solenodon paradoxus PI 900 4262.00 478.00 8.36 6.17 6.80 Talpa europaea PI 76 953.00 60.00 6.86 4.09 4.33 Tupaia glis PI 150 2959.00 128.00 7.99 4.85 5.01 Tupaia minor PI 70 2430.00 94.30 7.80 4.55 4.25 Urogale evere tti PI 275 3997.00 186.00 8.29 5.23 5.62 Avahi laniger S 860 9075.00 80.60 9.11 4.39 6.76 Cheirogaleus major S 450 6323.00 155.00 8.75 5.04 6.11 Cheirogaleus medius S 177 2941.00 99.30 7.99 4.60 5.18 Daubentonia madagascariensis S 2800 42611.00 693.00 10.66 6.54 7.94 Galago crassicaudatus S 850 9602.00 180.00 9.17 5.19 6.75 Galago demidovii S 81 3203.00 84.40 8.07 4.44 4.39 Galago senegalensis S 186 4512.00 81.80 8.41 4.40 5.23 Hapalemur simus S 1300 8868.00 79.40 9.09 4.37 7.17 Indri indri S 6250 36159.00 142.00 10.50 4.96 8.74 Lemur fulvus S 1400 22053.00 229.00 10.00 5.43 7.24 Lemur variegatus S 3000 29713.00 374.00 10.30 5.92 8.01 Lepilemur ruficaudatus S 915 7167.00 131.00 8.88 4.88 6.82 Loris gracilis S 322 6269.00 88.10 8.74 4.48 5.77 Mi crocebus murinus S 54 1663.00 40.30 7.42 3.70 3.99 Nycticebus cougang S 800 11755.00 164.00 9.37 5.10 6.68 Perodicticus potto S 1150 13212.00 312.00 9.49 5.74 7.05 Propithecus verreauxi S 3480 25080.00 168.00 10.13 5.12 8.15 Data from Stephan et al. (1 sensu Stephan et al. (1970); BM= body mass in grams (g); Cat.= category; ECV= endocranial volume in cubic millimeters (mm 3 ); H= haplorhine primates; ln BM= natural log of BM; ln ECV= natural log of ECV; ln OBV= natural log of OBV; OBV= olfactory bulb volume in cubic millimeters (mm 3 sensu Stephan et al. (1970); S= strepsirrhine primates

PAGE 105

10 5 APPENDIX B CODE Using the Body Mass Code from Gingerich (1990) Adapted for R In order to use the code from Gingerich (1990) adapted for the statistical program R (R Core Team, 2014) that is reported here, a comma delimited file (.csv) file must first be produced with the necessary formatting. This can be achieved by the use of the program Microsoft Excel. In M icrosoft Excel, open a new spreadsheet. Row 1 should be filled with these column titles: ulnal length; for metatarsal diameter. All measurements should be in millimeters (mm) and should be taken following protocols described in Alexander et al. (1979) and Gingerich (1990). Once these data are placed in the Microsoft Excel spreadsheet f ile, save the spreadsheet as a comma delimited (.csv) file in the appropriate directory folder that may be accessed by R. below. After sub setting the unique i column and selecting an output (export) file in the appropriate places within the code, this code will produce another comma delimited file with a table of body mass estimates for that specimen using t he equations from the original code in Gingerich (1990).

PAGE 106

106 Body Mass Code from Gingerich (1990) Adapted for R Copy and paste the following code into a script window of R: ##### BEGINNING OF CODE for Body Mass Equations Code translated from Gingerich, 1990## #### ######INSERT COMMA DELIMITED FILE NAME BELOW###### x< read.csv("INSERT NAME OF FILE HERE .csv") #####subset relevant categories PLACE UNIQUE SPECIMEN NAME BELOW ####### C< x[which(x$Catalog == 'INSERT UNIQUE SPECIMEN NAME HERE FROM Catalog COLUMN HERE' ),] #####column 2: body mass estimates##### HLe< 10^( 1.5579)*(C[,4])^(2.6752) HLR2< .9685 HLw< HLR2*log(HLe) ULe< 10^( 1.8459)*(C[,5])^(2.7162) ULR2< .97185 ULw< ULR2*log(ULe) MCLe< 10^( .063602)*(C[,6])^(2.4746) MCLR2< .91817 MCLw< MCLR2*log(MCLe) FLe< 10^( 1.7511)*(C[,7])^(2.6544) FLR2< .96504 FLw< FLR2*log(FLe) TLe< 10^( 2.6904)*(C[,8])^(3.0581) TLR2< .96384 TLw< TLR2*log(TLe) MTLe< 10^( 1.3562)*(C[,9])^(3.0604) MTLR2< .92179 MTLw< MTLR2*log(MTLe) HDe< 10^(1.2061)*(C[,10])^(2.5984) HDR2< .9 9246 HDw< HDR2*log(HDe) UDe< 10^(NA)*(C[,11])^(NA) UDR2< NA UDw< NA MCDe< 10^(2.1836)*(C[,12])^(2.7377) MCDR2< .98534 MCDw< MCDR2*log(MCDe) FDe< 10^(1.0632)*(C[,13])^(2.7418) FDR2< .99138 FDw< FDR2*log(FDe) TDe< 10^(1.1929)*(C[,14])^(2.6828) TDR2< .98138 TDw< TDR2*log(TDe)

PAGE 107

107 MTDe< 10^(1.7879)*(C[,15])^(2.9932) MTDR2< .9661 MTDw< MTDR2*log(MTDe) ntot< (HLw+ULw+MCLw+FLw+TLw+MTLw+HDw+MCDw+FDw+TDw+MTDw) dtot< (HLR2+ULR2+MCLR2+FLR2+TLR2+MTLR2+HDR2+MCDR2+FDR2+TDR2+MTDR2) wa< exp(ntot/dtot) ######column 3 &4: pred iction interval min & max ##### HLt< 2.021 HLx< 1.9154 HLs2yx< .06655 HLsb< 0.077259 HLn< 41 ULt< 2.022 ULx< 1.9787 ULs2yx< 0.05931 ULsb< .074985 ULn< 40 MCLt< 2.052 MCLx< 1.3217 MCLs2yx< .14524 MCLsb< .14218 MCLn< 29 FLt< 2.022 FLx< 2.0104 FLs2 yx< 0.075244 FLsb< .081958 FLn< 40 TLt< 2.021 TLx< 2.0459 TLs2yx< 0.076397 TLsb< 0.094856 TLn< 41 MTLt< 2.048 MTLx< 1.4612 MTLs2yx< .15419 MTLsb< .16846 MTLn< 30 HDt< 2.021 HDx< .90825 HDs2yx< .015928

PAGE 108

108 HDsb< .036267 HDn< 41 UDt< 0 UDx< 0 UDs2yx< 0 UDsb< 0 UDn< 0 MCDt< 2.064 MCDx< .35153 MCDs2yx< .033483 MCDsb< .068174 MCDn< 26 FDt< 2.022 FDx< .91988 FDs2yx< .018552 FDsb< .041474 FDn< 40 TDt< 2.021 TDx< .8846 TDs2yx< .039327 TDsb< 0.059168 TDn< 41 MTDt< 2.06 MTDx< .45294 MTDs2yx< 0.074488 MTDsb< .1121 3 MTDn< 27 HLsy< sqrt(HLs2yx + (HLs2yx/HLn) + (HLsb^2 *(log(C[,4]) /log (10) HLx)^2)) HLhalf< abs(HLt*HLsy) HLlow< 10^(log(HLe)/log(10) HLhalf) HLhigh< 10^(log(HLe)/log(10)+HLhalf) ULsy< sqrt(ULs2yx + (ULs2yx/ULn) + (ULsb^2 *(log(C[,5]) /log (10) ULx)^ 2)) ULhalf< abs(ULt*ULsy) (ULlow< 10^(log(ULe)/log(10) ULhalf)) (ULhigh< 10^(log(ULe)/log(10)+ULhalf)) MCLsy< sqrt(MCLs2yx + (MCLs2yx/MCLn) + (MCLsb^2 *(log(C[,6]) /log(10) MCLx)^2))

PAGE 109

109 MCLhalf< abs(MCLt*MCLsy) (MCLlow< 10^(log(MCLe)/log(10) MCLhalf)) (MCLhi gh< 10^(log(MCLe)/log(10)+MCLhalf)) FLsy< sqrt(FLs2yx + (FLs2yx/FLn) + (FLsb^2 *(log(C[,7]) /log (10) FLx)^2)) FLhalf< abs(FLt*FLsy) (FLlow< 10^(log(FLe)/log(10) FLhalf)) (FLhigh< 10^(log(FLe)/log(10)+FLhalf)) TLsy< sqrt(TLs2yx + (TLs2yx/TLn) + (TLsb^2 *(log(C[,8]) /log (10) TLx)^2)) TLhalf< abs(TLt*TLsy) (TLlow< 10^(log(TLe)/log(10) TLhalf)) (TLhigh< 10^(log(TLe)/log(10)+TLhalf)) MTLsy< sqrt(MTLs2yx + (MTLs2yx/MTLn) + (MTLsb^2 *(log(C[,9])/log (10) MTLx)^2)) MTLhalf< abs(MTLt*MTLsy) (MTLlow< 10^(log(M TLe)/log(10) MTLhalf)) (MTLhigh< 10^(log(MTLe)/log(10)+MTLhalf)) HDsy< sqrt(HDs2yx + (HDs2yx/HDn) + (HDsb^2 *(log(C[,10]) /log (10) HDx)^2)) HDhalf< abs(HDt*HDsy) (HDlow< 10^(log(HDe)/log(10) HDhalf)) (HDhigh< 10^(log(HDe)/log(10)+HDhalf)) UDsy< sqrt(UD s2yx + (UDs2yx/UDn) + (UDsb^2 *(log(C[,11]) /log (10) UDx)^2)) UDhalf< abs(UDt*UDsy) (UDlow< 10^(log(UDe)/log(10) UDhalf)) (UDhigh< 10^(log(UDe)/log(10)+UDhalf)) MCDsy< sqrt(MCDs2yx + (MCDs2yx/MCDn) + (MCDsb^2 *(log(C[,12]) /log (10) MCDx)^2)) MCDhalf< a bs(MCDt*MCDsy) (MCDlow< 10^(log(MCDe)/log(10) MCDhalf)) (MCDhigh< 10^(log(MCDe)/log(10)+MCDhalf)) FDsy< sqrt(FDs2yx + (FDs2yx/FDn) + (FDsb^2 *(log(C[,13]) /log (10) FDx)^2)) FDhalf< abs(FDt*FDsy) (FDlow< 10^(log(FDe)/log(10) FDhalf)) (FDhigh< 10^(log(FDe )/log(10)+FDhalf))

PAGE 110

110 TDsy< sqrt(TDs2yx + (TDs2yx/TDn) + (TDsb^2 *(log(C[,14]) /log (10) TDx)^2)) TDhalf< abs(TDt*TDsy) (TDlow< 10^(log(TDe)/log(10) TDhalf)) (TDhigh< 10^(log(TDe)/log(10)+TDhalf)) MTDsy< sqrt(MTDs2yx + (MTDs2yx/MTDn) + (MTDsb^2 *(log(C[,15 ]) /log (10) MTDx)^2)) MTDhalf< abs(MTDt*MTDsy) (MTDlow< 10^(log(MTDe)/log(10) MTDhalf)) (MTDhigh< 10^(log(MTDe)/log(10)+MTDhalf)) wa2< max(c(HLlow,ULlow, MCLlow,FLlow,TLlow,MTLlow,HDlow,MCDlow,FDlow,TDlow,MTDlow)) wa3< min(c(HLhigh,ULhigh,MCLhigh,FLhig h,TLhigh,MTLhigh,HDhigh,MCDhigh ,FDhigh,TDhigh, MTDhigh)) ####multiple regression equations################ ###length and diameter, all species### HL1< (1.1263)*(log(C[,4])/log(10)) UL1< ( .34705)*(log(C[,5])/log(10)) MCL1< ( .42902)*(log(C[,6])/log(10) ) FL1< (.14743)*(log(C[,7])/log(10)) TL1< ( .51599)*(log(C[,8])/log(10)) MTL1< (.41567)*(log(C[,9])/log(10)) HD1< (.67837)*(log(C[,10])/log(10)) UD1< (log(C[,11])/log(10)) MCD1< (.83146)*(log(C[,12])/log(10)) FD1< (.97864)*(log(C[,13])/log(10)) TD1< ( .277 35)*(log(C[,14])/log(10)) MTD1< (.056296)*(log(C[,15])/log(10)) MR1< 10^(1.0539+HL1+UL1+MCL1+FL1+TL1+MTL1+HD1+MCD1+FD1+TD1+MTD1) MR1 ###length and diameter, artiodactyls removed### HL2< (1.369)*(log(C[,4])/log(10)) UL2< ( .53119)*(log(C[,5])/log(10)) MC L2< ( .065223)*(log(C[,6])/log(10)) FL2< (.87466)*(log(C[,7])/log(10)) TL2< ( 1.0127)*(log(C[,8])/log(10))

PAGE 111

111 MTL2< (.13456)*(log(C[,9])/log(10)) HD2< (.6123)*(log(C[,10])/log(10)) UD2< (log(C[,11])/log(10)) MCD2< (.90372)*(log(C[,12])/log(10)) FD2< (.73021)* (log(C[,13])/log(10)) TD2< ( .46398)*(log(C[,14])/log(10)) MTD2< ( .098183)*(log(C[,15])/log(10)) MR2< 10^(.90053+HL2+UL2+MCL2+FL2+TL2+MTL2+HD2+MCD2+FD2+TD2+MTD2) MR2 ###length only, all species ###### HL3< (1.1494)*(log(C[,4])/log(10)) UL3< (.54825)*(l og(C[,5])/log(10)) MCL3< ( .015956)*(log(C[,6])/log(10)) FL3< (1.6607)*(log(C[,7])/log(10)) TL3< ( 1.7635)*(log(C[,8])/log(10)) MTL3< (.93805)*(log(C[,9])/log(10)) MR3< 10^( 1.0061+HL3+UL3+MCL3+FL3+TL3+MTL3) MR3 ###length only, artiodactyls removed##### HL4< (2.0805)*(log(C[,4])/log(10)) UL4< ( .82606)*(log(C[,5])/log(10)) MCL4< (.16526)*(log(C[,6])/log(10)) FL4< (1.6647)*(log(C[,7])/log(10)) TL4< ( 1.1229)*(log(C[,8])/log(10)) MTL4< (.3828)*(log(C[,9])/log(10)) MR4< 10^( .81709+HL4+UL4+MCL4+FL4+TL4+MT L4) MR4 ######place in a table, export to .csv####### (col1< rbind(C[,4],C[,5],C[,6],C[,7],C[,8],C[,9],C[,10],C[,11],C[,12], C[,13],C[,14],C[,15],11,"11 L&D",NA)) col2< rbind(HLe, ULe, MCLe, FLe, TLe, MTLe, HDe, UDe, MCDe, FDe, TDe, MTDe,wa,MR1,MR2) col 3< rbind(HLlow,ULlow, MCLlow,FLlow,TLlow,MTLlow,HDlow,UDlow,MCDlow,FDlow,TDlow,MTDlow, wa2, "6 L",NA)

PAGE 112

112 col4< rbind(HLhigh,ULhigh,MCLhigh,FLhigh,TLhigh,MTLhigh,HDhigh,UDhigh, MCDhigh,FDhigh,TDhigh, MTDhigh,wa3,MR3,MR4) table1< cbind(col1,col2,col3,col4) col names(table1)< c("Measurement (mm)", "Predicted body mass(g)", "Minimum (g)", "Maximum(g)") rownames(table1)< c("Humerus length", "Ulna length", "Metacarpal length", "Femur length", "Tibia length", "Metatarsal length", "Humerus diameter", "Ulna diameter", "Metacarpal diameter", "Femur diameter", "Tibia diameter", "Metatarsal diameter", "N, geometric mean, min, max", "Multiple regression (all species)", "Multiple regression (art iod. removed)") ######INSERT NAME OF OUTPUT FILE FILE YOU WOULD LIKE TO EXP ORT TABLE TO BELOW##### write.csv(table1," NAME OF OUTPUT FILE .csv") ###END OF CODE########## Body Mass Equations from Boyer (2009) Adapted for R Likewise, the same comma delimited file that was prepared for use with the previous code may be used with thi s code as well. The original equations used in t his code appeared in Boyer (2009 ). This code will also produce an output file with a table of body masses for a specimen of interest. Copy and paste the following code into a script window of R: ##### BEGINNI NG OF CODE for Body Mass Equations from Boyer 2009 (Dissertation) ##### INSERT COMMA DELIMITED FILE NAME BELOW###### x< read.csv("INSERT NAME OF FILE HERE .csv") #####subset relevant categories PLACE UNIQUE SPECIMEN NAME BELOW ####### C< x[which(x$Catalog = = 'INSERT UNIQUE SPECIMEN NAME HERE FROM Catalog COLUMN HERE'),] #####col umn 2: body mass estimates##### e< 2.71828182846 ln < function(x) { log(x, 2.71828182846)} (HLe< e^( 2.9741)*(C[,4])^(2.3405))

PAGE 113

113 HLR2< .9104 (HLw< HLR2*ln(HLe)) ULe< e^( 3.2743)*(C [,5])^(2.3643) ULR2< .9008 ULw< ULR2*ln(ULe) MCLe< e^(1.3759)*(C[,6])^(2.0101) MCLR2< .8585 MCLw< MCLR2*ln(MCLe) FLe< e^( 6.2301)*(C[,7])^(2.9254) FLR2< .8902 FLw< FLR2*ln(FLe) TLe< e^( 7.2873)*(C[,8])^(3.1823) TLR2< .8889 TLw< TLR2*ln(TLe) MTLe< e ^( 1.4216)*(C[,9])^(2.6293) MTLR2< .7348 MTLw< MTLR2*ln(MTLe) HDe< e^(2.8475)*(C[,10])^(2.4992) HDR2< .9451 HDw< HDR2*ln(HDe) UDe< e^(3.8611)*(C[,11])^(2.3251) UDR2< 0.901 UDw< UDR2*ln(UDe) MCDe< e^(5.5167)*(C[,12])^(2.421) MCDR2< .8919 MCDw< MCDR2 *ln(MCDe) FDe< e^(2.4022)*(C[,13])^(2.6143) FDR2< .9028 FDw< FDR2*ln(FDe) TDe< e^(3.8611)*(C[,14])^(2.3251) TDR2< .901 TDw< TDR2*ln(TDe) MTDe< e^(5.0395)*(C[,15])^(2.7378) MTDR2< .8938 MTDw< MTDR2*ln(MTDe) ntot< (HLw+ULw+MCLw+FLw+TLw+MTLw+HDw+UDw+MC Dw+FDw+TDw+MTDw) dtot< (HLR2+ULR2+MCLR2+FLR2+TLR2+MTLR2+HDR2+UDR2+MCDR2+FDR2+TDR2+MTDR2 ) wa< exp(ntot/dtot) col1< rbind(C[,4],C[,5],C[,6],C[,7],C[,8],C[,9],C[,10],C[,11],C[,12], C[,13],C[,14],C[,15],12) col2< rbind(HLe, ULe, MCLe, FLe, TLe, MTLe, HDe, UD e, MCDe, FDe, TDe, MTDe,wa)

PAGE 114

114 table1< cbind(col1,col2) rownames(table1)< c("Humerus length", "Ulna length", "Metacarpal length", "Femur length", "Tibia length", "Metatarsal length", "Humerus diameter", "Ulna diameter", "Metacarpal diameter", "Femur diame ter", "Tibia diameter", "Metatarsal diameter", "N, geometric mean, min, max") colnames(table1)< c("Measurement (mm)", "Predicted body mass(g)") table1 ######INSERT NAME OF OUTPUT FILE FILE YOU WOULD LIKE TO EXPORT TABLE TO BELOW##### write.csv(table1," NA ME OF OUTPUT FILE .csv") ###END OF CODE##########

PAGE 115

115 LIST OF REFERENCES Alexander, J.P., 1994. Sexual dimorphism in notharctid primates. Folia Primatol. 63, 59 62. Alexander, J.P., Burger, B.J., 2001. Stratigraphy and taphonomy of Grizzly Buttes, Bridger formation, and the middle Eocene of Wyoming. In Gunnell, G. F. (Ed.), Eocene Biodiversity: Unusual Occurrences and Rarely Sampled Habitats. Plenum Press, New York, pp. 165 196. Barton, R.A., 2004. Binocularity and brain evolution in primates. Proc. Nat. A cad. Sci. USA 101, 10113 10115. Alexander, R.M., Jayes, A.S., Maloiy, G.M.O., Wathuta, E.M., 1979. Allometry of the limb bones of mammals rom shrews ( Sorex ) to elephant ( Loxodonta ). J. Zool. 189, 305 314. Beard, C.K., Dagosto, M., Gebo, D.L., Godinot, M. , 1986. Interrelationships among primate higher taxa. Nature 331, 712 714. Beard, K.C., 1988. New notharctine primate f osils from the early Eocene of New Mexico and southern Wyoming and he phylogeny of Notharctinae. Am. J. Phys. Anthropol. 75, 439 469. B loch, J.I., Silcox, M.T., 2001. New basicrania of Paleocene Eocene Ignacius : re evaluation of the plesiadapiform dermopteran link. Am. J. Phys. Anthropol. 116, 184 198. Bloch, J.I., Silcox, M.T., Boyer, D.M., Sargis, E. J., 2007. New Paleocene skeletons and the relationship of plesiadapiforms to crown clade primates. Proc. Nat. Acad. Sci. USA 104, 1159 1164. Boyer, D.M., 2009. New cranial and postcranial remain of late Paleocene Plesiadapidae description and evolutionary implications. Ph.D. dissertation, Stony Book University. Boyer, D.M., Seifert, E.R., Gladman, J.T. Bloch, J.I., 2013a. Evolution and allometry of calcaneal elongation in living and extinct primates. PLOS One 8, e67792 Boye r, D.M., Yapuncich, G.S., Chester, S.G.B., Bloch, J.I., Godinot, M., 2013 b . Hands of early primates. Am. J. Phys. Anthropol. 57, 33 78. Cartmill, M., 1992. New views on primate origins. Evol. Anthropol. 1, 105 111. Conroy, G.C., 1987. Problems in body wei ght estimation in fossil primates. Int. J. Primatol. 8, 115 137. Covert, H.H., 1985. Adaptations and evolutionary relationships of the Eocene Primate family Notharctidae. Ph.D. dissertation, Duke University. Covert, H.H., 1986. The biology of early Ceno zoic primates. In Swindler, D.R., Erwin, J. (Eds.), Comparative Primate Biology vol. 1: Systematics, Evolution, and Anatomy. Alan R. Liss, New York.

PAGE 116

116 Covert, H.H. 1990. Phylogenetic relationships among the Notharctinae of North America. Am. J. Phys. Anthr opol.81, 381 397. Covert, H.H.,1995. Locomotor adaptations of Eocene primates: adaptive diversity among the earliest prosimians. In Alterman, L., Doyle, G.A., Izard, M.K. (Eds.), Creatures of the Dark: The Nocturnal Prosimians. Plenum Press, New York. C onroy, G.C., 1987. Problems of body weight estimation in fossil primates. Int. J. Primatol. 8, 115 137. Clyde, W.C., Zonnveld, J. P., Stametakos, J., Gunnell, G.F., Bartels, W.S., 1997. Magnetostratigraphy across the Wasatchian Bridgerian (early to middle Eocene) in the western Green River Basin, Wyoming. J. Geol. 105, 657 669. Clyde, W.C., Sheldon, N.D., Koch, P.L., Gunnell, G.F., Bartels, W.S., 2001. Linking the Wasatchian/Bridgerian boundary to the Cenozoic global climate optimum: new magnetostratigra phic and isotopic results from South Pass, Wyoming. Palaeogeogr. Palaeoclimatol. Palaeoecol. 167, 175 199. Dagosto, M., Terranova, C.J., 1992. Estimating the body size of Eocene primates: a comparison of results from dental and postcranial variables. Int . J. Primatol. 13, 307 344. Dunbar, R.I.M., Shultz, S., 2007. Understanding primate brain evolution. Phil. Trans. R. Soc. B 362, 649 658. Dunkle, D.H., 1959. U.S. National Museum. Sc. Vert. Paleo. News Bull. 56, 11. Eisenberg, J.F., 1981. The Mammalian Radiations: an Analysis of the Trends in Evolution, Adaptation, and Behavior. Univ. Chicago Press, Chicago and London. Edinger, Tilly, 1964. Midbrain exposure and overlap in mammals. Am. Zool. 1, 5 19. Elliot Smith, G., 1902. On the morphology of the br ain in Mammalia, with special reference to that of lemurs, recent and extinct. Trans. Linn. Soc. Lond. (Zool.) 8, 319 432. Fleagle, J.G., 1999. Primate Adaptation and Evolution. Academic Press, New York. Fleagle, J.G., 2013. Primate Adaptation and Evolu tion. Academic Press, New York. Franzen, J.L., Gingerich, P.D., Habersetzer, J., Hurum, J.H., Koeningswald, W.v., Smith, B.H., 2009. Complete primate skeleton from the middle Eocene of Messel in Germany: Morphology and Paleobiology. PLoS One 4,1 27. Gaz in, C.L., 1958. A Review of the Middle and Upper Eocene Primates of North America (with 14 Plates) . Smithsonian Misc. Coll. 136, 1 116.

PAGE 117

117 Gazin, C.L., 1965. An endocranial cast of the the Bridger Middle Eocene primate Smilodectes gracilis . Smithsonian Misc. Coll. 149, 1 14. Gilbert, C.C., 2005. Dietary ecospace and the diversity of euprimates during the Early and Middle Eocene. Am. J. Phys. Anthropol. 126, 237 249. Gingerich, P.D., Martin, R.D., 1981 . Cranial morphology and adaptations in Eocene Adapidae I I: the Crambridge skull of Adapis parisiensis . Am. J. Phys. Anthropol. 56, 235 257. Gingerich, P.D., Smith, B.H., Rosenberg, K.R., 1982. Allometric scaling in the dentition of primates and prediction of body weight from tooth size in fossils. Am. J. Phys. Anthropol. 58, 81 100 Gingerich, P.D., 1990. Prediction of body mass in mammalian species from long bone lengths and diameters. Contrib. Mus. Paleontol. Univ. Mich. 28, 79 92. Gingerich, P.D., Gunnell, G.F., 2005. Brain of Plesiadapis cookei (Mammalia, Proprimates): surface morphology and encephalization compared to those of Primates and Dermoptera. Contrib. Mus. Paleontol. Univ. Mich. 31, 185 195. Gingerich, P.D., Franzen, J.L., Habersetzer, J., Hurum, J.H., Smith, B.H., 2010. Darwinius masillae is a H aplorhine reply to Williams et al. (2010). J. Hum. Evol. 59, 574 579. Gregory, W.K., 1920. On the structure and relations of Notharctus , an American Eocene primate. Mem. Am. Mus. Nat. Hist., new ser. 3, 49 243 Gingerich, P.D., Smith, B.H., 2010. Premolar development and eruption in the early Eocene adapoids Cantius ralstoni and Cantius abditus (Mammalia, Primates). Contrib. Mus. Paleontol. Univ. Mich. 32, 41 47. Grande, L., Lundberg, J.G., 1988. Revision and redescription of the genus Astephus (Silurifor mes: Ictaluridae) with a discussion of its phylogenetic relationships. J. Vert. Paleo. 8, 139 171. Gregory, W.K., 1920. On the structure and relations of Notharctus , an American Eocene primate. Mem. Am. Mus. Nat. Hist. 3, 49 243. Gunnell G.F. 1995. New N otharctinae (Primates, Adapiformes) skull from the Uintan (Middle Eocene) of San Diego County, California. Am. J. Phys. Anthropol. 98, 447 470. Gunnell., G.F., 1998. Mammalian fauna from the lower Bridger Formation (Bridger A, early middle Eocene) of the southern Green River Basin, Wyoming. Contrib. Mus. Paleontol. Univ. Mich. 30, 83 130. Gunnell., G.F., 2002. Notharctine primates (Adapiformes) from the early to middle Eocene (Wasatchian Bridgerian) of Wyoming: transitional species and the origins of Noth arctus and Smilodectes . J. Hum. Evol. 43, 353 380.

PAGE 118

118 Gunnell, G.F., Rose, K.D., Rasmussen, D.T.,, 2008. Euprimates . In Janis, C., Gunnell, G.F., and Uhen, M.D. (eds.), Evolution of Tertiary Mammals in North America, volume 2: small mammals, xenarthrans, an d marine mammals. Cambridge University Press, New York. Gunnell, G.F., Murphey, P.C., Stucky, R.K., Townshend, K.E.B., Robinson, P., Zonn eveld, J.P., Bartels, W.S., 2009 . Biostratigraphy and biochronology of the latest Wasatchian, Bridgerian, and Uintan Vertebrate Paleontology, and biostratigraphy in honor of Michael O. Woodburne. Mus. N. Ariz. Bull. 65, 279 330. is, University of Kansas. Gurche, J.A., 1982. Early primate brain evolution. In: Armstrong, E., Falk, D. (Eds.), Primate Brain Evolution: Methods and Concepts. Plenum, New York, pp 227 246. Hamrick, M.W., Alexander, J.P., 1996. The hand skeleton of Noth arctus tenebrosus (Primates, Notharctidae) and its significance for the origin of the primate hand. Am. Mus. Novit. 3182, 1 20. Harvey, P.H., Clutton Brock, T.H., Mace, G.M., 1980. Brain size and ecology in small mammals and primates. Proc. Nat. Acad. Sci . USA 77, 4387 4389. Heritage, S., 2014. Modeling olfactory bulb evolution through primate phylogeny. PLOS One 9, e113904. Hofer, H.O., 1962. Über die interpretation der ältesten fossilen primatengehirne. Bibliogr. Primatol. 5, 1 31. Hofer, H.O., Wilson , J.A., 1967. An endocranial cast of an early Oligocene primate. Folia Primatol. 5, 148 152. Hürzeler, J., 1948. Zur Stammegeschichte der Necrolemuriden. Schweiz. Palaont. Abh. 66, 3 46. Isler, K, Kirk, C.E., Miller, J.M.A., Albrecht, G.A., Gelvin, B.R. , Martin, R.D., 2008. Endocranial volumes of primate species: scaling analyses using comprehensive and reliable data set. J. Hum. Evol 55, 967 978. Jerison, H.J., 1973. Evolution of the Brain and Intelligence. Academic Press, New York. Jerison, H.J., 197 9. Brain, body, and encephalization in early primates. J. Hum. Evol. 8, 615 613. Kay, R.F., Ross, C., Williams, B.A., 1997. Anthropoid origins. Science 275, 797 804 Kay, R.F., Kirk, E.C., 2000. Osteological evidence for the evolution of activity pattern and visual acuity in primates. Am. J. Phys. Anthropol. 113, 235 262.

PAGE 119

119 Kirk, E.C., Kay, R. F., 2004. The evolution of high visual acuity in the Anthropoidea. In Ross, C.F., Kay, R.F. (Eds.), Anthropoid Origins: New Visions. Plenum Press, New York, pp. 539 6 02. Kirk, E.C., 2006a. Visual influences on primate encephalization. J. Hum. Evol. 51, 76 90. Kirk, E.C., 2006b. Effects of activity pattern on eye size and orbital aperture size in primates. J. Hum. Evol. 51, 159 170. Kirk, E.C., Daghighi, P., Macri ni, T.E., Bhullar, B.A.S., Rowe, T.B., 2014. Cranial anatomy of the Duchesnean primate Rooneyia viejaensis : new insights from high resolution computed tomography. J. Hum. Evol. 74., 82 95. Lay, M. Fossil collecting expeditions led by the division of verte brate paleontology during the http://www.paleobiology.si.edu/pdfs/The%201950's%20Collections.pdf > Le Gros Clark, W.E., 1945. Note on the paleontology of the lemur oid brain. J. Anat. 79, 123 126. Long, J.O., Cooper, R.W., 1968. Physical growth and dental eruption in captive bred squirrel monkeys, Saimiri sciurus (Leticia, Columbia). In: Rosenblum, L., Cooper, R. (eds.), The Squirrel Monkey. Academic Press, New Yor k and London, pp. 169 205. Macrini, T.E., Rowe, T., VandeBerg, J.L., 2007 . Cranial endocasts from a growth series of Monodeplphis domestica (Didelphidae, Marsupialia): a study of individual and ontogenetic variation. Maiolino, S., Boyer, D.M., Bloch, J. I., Gilbert, C.C., Groenke, J, 2012. Evidence for a grooming claw in a North American adapiform primate: implications for anthropoid origins. PLOS One 7, 1 28. Manocha, S.L., 1979. Physical growth and brain development of captive bred male and female squ irrel monkeys, Saimiri sciurus . Experientia 35, 96 98. Martin, R .D., 1990. Primate Origins and Evolution: a Phylogenetic Reconstruction. Chapman and Hall, London. Montgomery, S.H., Capellini, I., Barton, R.A., Mundy, N.I., 2010. Reconstructing the ups a nd downs of primate brain evolution: implications for adaptive hypotheses and Homo florisiensis . BMC Biol. 8, 9. Murphey, P.C., Lester, A., Bohor, B., Robinson, P., Evanoff, E., Larson, E., 1999. 40 Ar/ 39 Ar dating of volcanic ash deposits in the Bridger Fo rmation (middle Eocene), southwestern Wyoming. Geol. Soc. Am. Abstracts with Programs 33, A 233.

PAGE 120

120 Neumayer, L., 1906. Ueber das Gehirn von Adapis parisiensis Cuv. Cuv. Neues. Jahrb. Min. Geol. Palaont. 2, 100 104. Orliac, M.J., Ladevèze, S., Gingerich, P.D., Lebrun, R., Smith, T., 2014. Endocranial morphology of Paleocene Plesiadapis tricuspidens and the evolu tion of the early primate brain. Proc. R. Soc. B 281, 20132792. Plavcan, J.M., 2000. Inferring social behavior from sexual dimorphism in the fossil record. J. Hum. Evol. 39, 327 344, Plavcan, J.M., 2002. Sexual dimorphism in primate evolution. Yearb. Phy s. Anthropol. 44, 25 53 Plavcan, J.M., 2004. Evidence for Early Anthropoid Social Behavior. In Ross, C.F., Kay, R.F. (Eds.), Anthropoid Origins: New Visions. Plenum Press, New York, pp. 383 413. R Core Team 2015. R: a language and environment for statis tical computing. R Foundation for Statistical Computing, Vienna. Radinsky, L.B., 1967. The oldest primate endocast. Am. J. Phys. Anthropol. 27, 385 388. Radinsky, L.B., 1970. The fossil evidence of prosimian brain evolution. In: Noback, C.R., and Motagn a, W. (Eds.), The Primate Brain: Advances in Primatology, vol. 1. Appleton Century Crofts, New York, pp. 209 224. Radinsky, L.B., 1974. Prosimian brain morphology: functional and phylogenetic implications. In Martin, R.D., Doyle, G.A., Walker, A.C. (Eds.) , Research Seminar on Prosimian Biology 1972: London University. University of Pittsburg Press, Pittsburg. Radinsky, L.B., 1975. Primate brain evolution. Am. Sci. 63, 656 663. Radinsky, L.B., 1977. Early primate brains: facts and fiction. J. Hum. Evol. 6 , 79 86. Radinsky, L.B., 1978. Evolution of brain size in carnivores and ungulates. Am. Nat. 112, 815 831. Robinson, P., Gunnell, G.F., Walsh, S.L., Clyde, W.C., Storer, J.E., Stucky, R.K., Froehlich, D.J., Ferrusquia Villafranca, I., McKenna, M.C., 20 04. Wasatchian through Duschesnean biochronology. In Woodburn, M.O. (Ed.), Late Cretaceous and Cenozoic Mammals of North America. Columbia University Press, New York. Rose, K.D., Bown, T.M., 1991. Additional fossil evidence on the differentiation of the earliest euprimates. Proc. Nat. Acad. Sci. USA 88, 98 101. Rose, K.D., 1994. The earliest primates. Evol. Anthropol. 3, 159 173.

PAGE 121

121 Rose, K.D., MacPhee, R.D.E., Alexander, J.P., 1999. Skull of early Eocene Cantius adbitus (Primates: Adapiformes) and its ph ylogenetic implications, with a reevaluation of . Am. J. Phys. Anthropol. 109, 523 539. Sauther, M.L., Cuozzo, F.P., 2012. Understanding Eocene primate paleobiology using a comprehensive analysis of living primate ecology, biology, a nd behaviour. Paleobio. Palaeoenv. 92, 573 583. Schillaci, M.A., 2006. Sexual selection and the evolution of brain size in Primates. Plos One 1, e62. Schillaci, M.A., 2008. Primate mating systems and the evolution of neocortex size. J. Mammol. 89, 58 63. Sei f fert, E.R., Perry, J.M., Simons, E.L., Boyer, D.M., 2009. Convergent evolution of anthropoid like adaptations in Eocene adapiform primates. Nature 461, 1118 1122. Shultz, S., Dunbar, R., 2010. Encephaliation is not a universal macroevolutionary phe nomenon in mammals but is associated with sociality. Proc. Nat. Acad. Sci. USA 107, 21582 21586. Silcox, M.T., Bloch, J.I., Boyer, D.M., Godinot, M., Ryan, T.M., Spoor, F., Walker, A., 2009a. Semicircular canal system in early primates. J. Hum. Evol. 56, 315 327. Silcox, M.T., Dalmyn, C.K., Bloch, J.I., 2009b. Virtual endocast of Ignacius graybullianus (Paromyidae, Primates) and brain evolution in early primates. Proc. Nat. Acad. Sci. USA 106, 10987 10992. Silcox, M.T., Benham, A.E., Bloch, J.I., 2010. Endocasts of Microsyops (Microsyopidae, Primates), and the evolution of the brain in primitive primates. J. Hum. Evol. 58, 505 521. Silcox, M.T., Dalmyn, C.K., Hrenchuk, A., Bloch, J.I., Boyer, D.M., Houde, P., 2011. Endocranial morphology of Labidolemu r kayi (Apatemyidae, Apatotheria) and its relevance to the study of brain evolution in euarchontoglires. J. Vert. Paleo. 31, 1314 1325. Simons, E.L., Ramussen, D.T., 1996. The skull of Catopithecus browni , an early Tertiary catarrhine. Am. J. Phys. Anthr opol. 100, 261 292. Smith, B.H., 1991. Age of weaning approximates emergence of the first permanent molar in non human primates. Am. J. Phys. Anthropol. 12, 163 164. Smith, B.H., Crummett. T.L., Brandt, K.L., 1994. Ages of eruption of primate teeth: a compendium for aging individuals and comparing life histories. Yearb. Phys. Anthropol. 37, 177 231. Smith, R.J., 1993. Bias in equations used to estimate fossil primate body mass. J. Hum. Evol. 25, 31 41.

PAGE 122

122 Smith, R.J., 1994. Regression models for predicti on equations. J. Hum. Evol. 26, 239 244. Steiper, M.E., Seiffert, E.R., 2012. Evidence for a convergent slowdown in primate molecular rates and its implications for the timing of primate evolution. Proc. Nat. Acad. Sci. 6006 6011 Stephan, H., Bauchot., R ., Andy, O.J., 1970. Data on size of the brain and of various brain parts in insectivores and primates. In: Noback, C.R., Montagna, W. (Eds.), The Primate Brain, Appleton Century Crofts, New York, pp. 289 297. Szalay, F.S., 1969. Mixodectidae. Microsyopi dae, and the insectivore primate transition. Bull. Am. Mus. Nat. Hist. 140, 195 330. Szalay, F.S., Delson, E., 1979. Evolutionary History of the Primates, Academic Press, New York. Takai M., Shigehara, N., Egi, N., Tsubamoto, T., 2003. Endocranial cast and morphology of the olfactory bulb of Amphipithecus morgaungensis (latest middle Eocene of Myanmar). Primates 44, 137 144. Tattersall, I., Schwartz, J.H., 1974. Craniodental morphology and the systematics of the Malagasy lemurs (Primates, Prosimii). An thropol. Pap. Am. Mus. Nat. Hist. 52, 139 192. Williams, B.A., Kay, R.F., Kirk, C.C., 2010a. New Perspectives on anthropoid origins. Proc. Nat. Acad. Sci. USA 107, 4797 4804. Williams, B.A., Kay, R.F., Kirk, E.C., Ross, C.F., 2010b. Darwinius massilae i s a strepsirrhine a reply to Franzen et al. (2009). J. Hum. Evol. 59, 567 573. Yapuncich, G.S., Gladman, J.T., Boyer, D.M., In review. Estimating euarchontan body mass: a comparison of tarsal and dental variables. Am. J. Phys. Anthropol.

PAGE 123

123 BIOGRAPHICAL SKETCH Arianna R. Harrington earned her Bachelor of Arts with highest honors in a nthropology and her Bachelor of Science with high honors in z oology in the spring 2013 at the University of Florida. She entered graduate school in the Department of Biology at at th e University of Florida in f all 2013 , and graduated with a Master of Science degree in s pring 2015. She will be starting her Ph.D . in the Department of Evolutionary Anthropology at Duke University in f all 2015 .