THE JOURNAL OF COMPARATIVE NEUROLOGY 310:429-474 (1991)
Myelo- and Cytoarchitecture of the Granular Frontal Cortex and Surrounding Regions in the Strepsirhine Primate Galago and the Anthropoid Primate Macaca TODD M. PREUSS AND PATRICIA S. GOLDMAN-RAKIC Section of Neurobiology, Yale University School of Medicine, New Haven, Connecticut 06510 (T.M.P., P.S.G.-R.); Department of Anthropology, Yale University, New Haven, Connecticut 06520 (T.M.P.)
ABSTRACT As the first part of a comparative investigation of primate frontal cortex, we compared the frontal architectonic organization of Galago, a small-brained, strepsirhine (or "prosimian") primate, to that of an anthropoid primate, Macaca, by using myelin- and Nissl-stained material. We were able to distinguish many more areas in both taxa than have been recognized in most previous studies of the primate frontal lobe. In particular, we were able to subdivide many of the areas shown in the commonly cited architectonic map of Walker (J. Comp. Neurol. 73.59436, 1940). Delineation of areas was greatly facilitated by the use of the Gallyas technique for staining myelin. The areal organization of much of frontal cortex (specifically, the premotor, orbital, and medial regions) appears to be very similar in Galago and Macaca. In these regions, we were able to recognize the same complement of areas in both taxa, with few exceptions. In the granular frontal cortex (GFC), by contrast, we were able to distinguish about twice as many areas in Macma as in Galago. For most of the GFC areas of Galago, there are architectonically similar areas in Macaca; the areas shared by both taxa correspond mainly to the arcuate and superior areas of Macaca (i.e., the region encompassed by Walker's areas 45, 8A, and 8B). However, there are many additional, more rostral, areas in Macaca for which there are no obvious homologues in Galago. In particular, Galago lacks cortex resembling the distinctive, lightly myelinated cortex of the Macaca principal sulcus (Walker's area 46 and its subdivisions). Our results are difficult to reconcile with the view that frontal lobe organization varies little across taxa. Rather, they suggest that granular frontal cortex underwent considerable change during primate evolution, including the addition of new areas in anthropoids. Key words: prefrontal cortex, association cortex, higher cortical function, evolution, monkey, prosimian
Few issues in comparative neuroanatomy have been so contentious as the status of primate granular frontal cortex (GFC), a region noted for its involvement in higher-order cognitive functions (Damasio, '85; Fuster, '89; GoldmanRakic, '87; Milner and Petrides, '84; Stuss and Benson, '86). On the one hand, Brodmann ('09, '121, who conducted a broad comparative survey of mammalian cytoarchitecture, concluded that the GFC is rudimentary or absent in mammals other than primates, and that GFC underwent considerable expansion during primate evolution, with the concomitant addition of new areas. Sanides ('64, '70) and workers influenced by him (Pandya et al., '88) have also O
1991 WILEY-LISS, INC.
argued that frontal cortex acquired new areas during primate evolution. Others have strongly disputed Brodmann's conclusions about the evolutionary distinctiveness of primate GFC, emphasizing the similarities of frontal lobe connectivity across mammals. In a very influential paper, Rose and Woolsey ('48) pointed out that the thalamic mediodorsal nucleus (MD), which projects to GFC in pri-
Accepted May 3,1991. Dr. Preuss's current address is Department of Psychology, Vanderbilt University, 301 Wilson Hall, Nashville, TN 37240.
T.M. PREUSS AND P.S. GOLDMAN-RAKIC
430 Abbreuiutions amyg AON
AS ASd ASV CaS Cd cgs CgSa
c1 cs
DA FEF FPO FSa FSP GFC GrAD GrD GrPL GrV Ia Ia-p Idg Ig IG Ins 10s
IPS LCS LOS LS
amygdala anterior olfactory nucleus arcuate sulcus arcuate sulcus, dorsal limb arcuate sulcus, ventral limb calcarine sulcus caudate nucleus cingulate sulcus cingulate sulcus, anterior part claustrum central sulcus dopamine frontal eye field frontoparietal opercular area anterior frontal sulcus posterior frontal sulcus granular frontal cortex Galago anterodorsal granular frontal area Gulugo dorsal granular frontal area Galago posterolateral frontal area Galago ventral frontal area agranular insular area periallocorticaldivision of agranular insular cortex dysgranular insular area granular insular area induseum griseum insular cortex inferior occipital sulcus intraparietal sulcus lateral calcarine sulcus lateral orbital sulcus lateral (Sylvian)sulcus
mates, projects to nongranular cortex in other mammals. Akert (’64) extended this argument, suggesting that MDprojection cortex in nonprimates is homologous to primate GFC, despite the absence of a distinct internal granular layer. Subsequently, the term “prefrontal cortex” has come into general use to denote the MD-projection cortex of all mammals, including the granular frontal cortex of primates. (For discussions of this point, see Fuster, ’89; Kolb and Whishaw, ’85; Markowitsch and Pritzel, ’79J1 There continues to be support for the view that there are no essential differences among mammals in the connectional or functional organization of prefrontal cortex (e.g., Eichenbaum et al., ’83; Kolb, ’84; Kolb and Whishaw, ’85). It is noteworthy that the contending points of view regarding comparative frontal lobe anatomy were established prior to advent of many of the techniques now commonly employed in neuroanatomy, such as the horseradish peroxidase and autoradiographic methods for tracing connections. These and other technical innovations have greatly altered our understanding of primate GFC organization. At midcentury, primate GFC was regarded as a relatively homogeneous entity, envincing little regional architectonic or functional specialization (e.g., Bonin, ’48; Bonin and Bailey, ’47; Lashley and Clark, ’46).Evidence ‘The origin of the infelicitous term “prefrontal,” as applied to the cortex, has long been something of a mystery. Divac (’88) attributes its first use to Richard Owen, a contemporary of Darwin and one of the leading authorities on comparative vertebrate anatomy in his day (Russell, ’16). This suggests a derivation for “prefrontal cortex.” The lobes of the mammalian cerebrum are named for the bones which overlay them: occipital, parietal, temporal, and frontal. There is no prefrontal bone in modern mammals. There is a prefrontal bone in other vertebrates, however, as Owen no doubt appreciated. This bone lies rostral t o the frontal bone, occupying the anterior wall of the bony orbit in many vertebrate groups (see Romer, ’70, or similar texts). Thus, the prefrontal bone bears an anatomical relationship to the frontal bone comparable to that between the “silent” prefrontal cortex and the electricallyexcitable frontal h e . , precentral) cortex.
LUS MD MF MFc MFr MOS MPOS Oa-p OB OF0 OLS Orb
0s
OTrS OTS OTu Para PCS PFC Pir PL POS PrCO PS Put RcS RoS RS RSa RSP SPS STS
lunate sulcus mediodorsal thalamic nucleus medial frontal areas caudal division of medial frontal cortex rostral division of medial frontal cortex medial orbital sulcus medial parieto-occipital sulcus periallocortical division of agranular orbital cortex olfactory bulb orbitofrontal opercular area olfactory sulcus orbital cortex orbital sulcus transverse occipital sulcus occipito-temporalsulcus olfactory tubercle paralimbic cortex paracalcarine sulcus prefrontal cortex piriform cortex prelimbic area parieto-occipital sulcus precentral opercular area principal sulcus putamen retrocalcarine sulcus rostral sulcus rhinal sulcus rhinal sulcus, anterior part rhinal sulcus, posterior part splenial sulcus superior temporal sulcus
from modern studies, however, and in particular from connectional studies of macaque monkeys (e.g., Cavada and Goldman-Rakic, ’89; Preuss and Goldman-Rakic, ’89; Seltzer and Pandya, ’89) and studies of metabolic activity in the human brain (Roland, ’84; Roland et al., ’87), suggests a much greater degree of regional structural and functional parcellation in the GFC than was suspected by most earlier workers. This new perspective on GFC organization raises old issues concerning its evolution: Do all mammals possess the full complement of GFC areas present in large-brained anthropoid primates, such as macaques and humans? If not, which areas are present only in anthropoids and which are more widely distributed among mammals? In order to address evolutionary issues such as these, we have undertaken a comparative anatomical study of primate granular frontal cortex. Specifically, we have employed modern architectonic and connectional techniques to compare the structure of GFC in two kinds of primates: macaque monkeys (genus Macaca), a group of anthropoid primates closely related to humans, and galagos (also known as bushbabies; genus Galago),which are representatives of the strepsirhine (or “prosimian”) primates. Some comments about strepsirhine primates, and why they are appropriate as a comparison group for anthropoids, are in order. As illustrated in Figure 1, the order Primates has two major branches, the haplorhines and strepsirhines. The haplorhine branch consists mainly of the familiar anthropoid primates-that is, monkeys, apes, and humans. The strepsirhines consist of lemurs, lorises, and galagos. In comparative anatomy, the evolutionary specializations of one group of animals are assessed by comparing them to closely related groups. So, strepsirhines such as Galago have long been used as an “outgroup” in the study of anthropoid specializations, as well as to determine what biological characteristics are common to all primates. This strategy has been pursued both in biological anthropology
ARCHITECTURE OF PRIMATE GRANULAR FRONTAL CORTEX
~-~PEalBidaalB Lemurs
Lorises
Galagos
431
-4 Tarsiers
New World Monkeys
Apes & Humans
Old World Monkeys
-
Primates Fig. 1. Evolutionary relationships of the placental mammals and of the primates. The relationships of the major groups of Iiving primates, as shown above, are widely agreed upon by modern workers (see, e.g., Fleagle, '88; Martin, '90; Richard, '85; Szalay and Delson, '79). Primates are composed of two major groups, strepsirhines and hap-
lorhines. The living stepsirhines are a fairly conservative group both anatomically and behaviorally. The haplorhines consist of the tarsiers (a single genus of small, nocturnal animals) and the familiar anthropoid ("higher") primates.
(reviewed by Cartmill, '82) and in neurobiology (e.g., Allman, '82; Carlson, '85; Diamond et al., '85; Kaas, '83). Strepsirhines are particularly well suited for this comparative role: although they are not literally primitive primates (they are in fact a diverse group with their own set of specializations), they are nonetheless evolutionarily conservative. That is, they retain many anatomical features of the earliest primates (and of related non-primate mammals) which have been modified during the evolution of anthropoid primates (Fleagle, '88; Martin, '90). Furthermore, there is good reason to suppose that neural organization is conservative in strepsirhines relative to anthropoids. Strepsirhine brains are small, averaging about half the size of anthropoid brains for a given body weight (Jerison, '73). Also, electrophysiological mapping studies indicate that while strepsirhines possess many of the visual and somatosensory cortical areas which are found in anthropoids, anthropoids have additional representations not found in strepsirhines (Allman et al., '79; Kaas, '83). Based on these considerations, we thought it likely that by comparing the strepsirhine Galago and the anthropoid Macaca, we could identify anthropoid specializations of GFC organization. In addition, by identifying GFC characteristics which are shared by both major groups of primates (strepsirhines and anthropoids), we can better characterize the basic plan of primate frontal lobe organization, and so provide a firmer basis for comparing primates to other mammals. In this paper, we report the results of a comparative myelo- and cytoarchitectonic study of granular frontal cortex of Galago and Macaca. The specific goals of this study were to make an areal parcellation of the GFC in
Galago and Macaca, and an initial interpretation of the homologies among areas, based on similarities of architectonic structure and location within the cortical mantle. This interpretation will subsequently be tested with connectional information (Preuss and Goldman-Rakic, '91b). There have been many previous architectonic studies of primate GFC, in macaques and other anthropoids (e.g., Barbas and Pandya, '89; Bonin and Bailey, '47; Brodmann, '05, '09; Rosabel, '67; Walker, '40) as well as in galagos and related strepsirhines (Bonin, '45; Brodmann, '08, '09, '12; Le Gros Clark, '31; Sanides and Krishnamurti, '67; Zilles and Schleicher, '80; Zilles et al., '79a,b, '82). (The widely used cytoarchitectonic maps of Brodmann "091 and Walker "401 are shown in Fig. 2.) In our view, these studies do not provide an adequate comparative foundation for interpreting the evolution of primate GFC. Their major shortcoming is a lack of agreement regarding the number and location of areal subdivisions within the GFC, even for a single species like Macaca mulatta (compare, for example, Bonin and Bailey, '47;and Walker, '40).There are even disagreements regarding the precise extent and borders of the granular frontal cortex, in part because there is no sharp discontinuity between cortex which possesses an internal granular layer (IV)and cortex which lacks it. For instance, Roberts and Akert ('63) and Barbas and Pandya ('87) would include much of the frontal opercular cortex within the granular frontal region, because it has a layer IV1although the same cortex was included within the agranular, precentral region by Brodmann ('09) and Walker ('40). Clearly, in order to compare the areal organization of GFC of Galago and Macaca, it is necessary to delimit the GFC in a consistent manner in both taxa and to have an
T.M. PREUSS AND P.S. GOLDMAN-RAKIC
432
12
26
Cercopithecus
i0
Myelo- and Cytoarchitecture of the Granular Frontal Cortex and Surrounding Regions in the Strepsirhine Primate Galago and the Anthropoid Primate Macaca TODD M. PREUSS AND PATRICIA S. GOLDMAN-RAKIC Section of Neurobiology, Yale University School of Medicine, New Haven, Connecticut 06510 (T.M.P., P.S.G.-R.); Department of Anthropology, Yale University, New Haven, Connecticut 06520 (T.M.P.)
ABSTRACT As the first part of a comparative investigation of primate frontal cortex, we compared the frontal architectonic organization of Galago, a small-brained, strepsirhine (or "prosimian") primate, to that of an anthropoid primate, Macaca, by using myelin- and Nissl-stained material. We were able to distinguish many more areas in both taxa than have been recognized in most previous studies of the primate frontal lobe. In particular, we were able to subdivide many of the areas shown in the commonly cited architectonic map of Walker (J. Comp. Neurol. 73.59436, 1940). Delineation of areas was greatly facilitated by the use of the Gallyas technique for staining myelin. The areal organization of much of frontal cortex (specifically, the premotor, orbital, and medial regions) appears to be very similar in Galago and Macaca. In these regions, we were able to recognize the same complement of areas in both taxa, with few exceptions. In the granular frontal cortex (GFC), by contrast, we were able to distinguish about twice as many areas in Macma as in Galago. For most of the GFC areas of Galago, there are architectonically similar areas in Macaca; the areas shared by both taxa correspond mainly to the arcuate and superior areas of Macaca (i.e., the region encompassed by Walker's areas 45, 8A, and 8B). However, there are many additional, more rostral, areas in Macaca for which there are no obvious homologues in Galago. In particular, Galago lacks cortex resembling the distinctive, lightly myelinated cortex of the Macaca principal sulcus (Walker's area 46 and its subdivisions). Our results are difficult to reconcile with the view that frontal lobe organization varies little across taxa. Rather, they suggest that granular frontal cortex underwent considerable change during primate evolution, including the addition of new areas in anthropoids. Key words: prefrontal cortex, association cortex, higher cortical function, evolution, monkey, prosimian
Few issues in comparative neuroanatomy have been so contentious as the status of primate granular frontal cortex (GFC), a region noted for its involvement in higher-order cognitive functions (Damasio, '85; Fuster, '89; GoldmanRakic, '87; Milner and Petrides, '84; Stuss and Benson, '86). On the one hand, Brodmann ('09, '121, who conducted a broad comparative survey of mammalian cytoarchitecture, concluded that the GFC is rudimentary or absent in mammals other than primates, and that GFC underwent considerable expansion during primate evolution, with the concomitant addition of new areas. Sanides ('64, '70) and workers influenced by him (Pandya et al., '88) have also O
1991 WILEY-LISS, INC.
argued that frontal cortex acquired new areas during primate evolution. Others have strongly disputed Brodmann's conclusions about the evolutionary distinctiveness of primate GFC, emphasizing the similarities of frontal lobe connectivity across mammals. In a very influential paper, Rose and Woolsey ('48) pointed out that the thalamic mediodorsal nucleus (MD), which projects to GFC in pri-
Accepted May 3,1991. Dr. Preuss's current address is Department of Psychology, Vanderbilt University, 301 Wilson Hall, Nashville, TN 37240.
T.M. PREUSS AND P.S. GOLDMAN-RAKIC
430 Abbreuiutions amyg AON
AS ASd ASV CaS Cd cgs CgSa
c1 cs
DA FEF FPO FSa FSP GFC GrAD GrD GrPL GrV Ia Ia-p Idg Ig IG Ins 10s
IPS LCS LOS LS
amygdala anterior olfactory nucleus arcuate sulcus arcuate sulcus, dorsal limb arcuate sulcus, ventral limb calcarine sulcus caudate nucleus cingulate sulcus cingulate sulcus, anterior part claustrum central sulcus dopamine frontal eye field frontoparietal opercular area anterior frontal sulcus posterior frontal sulcus granular frontal cortex Galago anterodorsal granular frontal area Gulugo dorsal granular frontal area Galago posterolateral frontal area Galago ventral frontal area agranular insular area periallocorticaldivision of agranular insular cortex dysgranular insular area granular insular area induseum griseum insular cortex inferior occipital sulcus intraparietal sulcus lateral calcarine sulcus lateral orbital sulcus lateral (Sylvian)sulcus
mates, projects to nongranular cortex in other mammals. Akert (’64) extended this argument, suggesting that MDprojection cortex in nonprimates is homologous to primate GFC, despite the absence of a distinct internal granular layer. Subsequently, the term “prefrontal cortex” has come into general use to denote the MD-projection cortex of all mammals, including the granular frontal cortex of primates. (For discussions of this point, see Fuster, ’89; Kolb and Whishaw, ’85; Markowitsch and Pritzel, ’79J1 There continues to be support for the view that there are no essential differences among mammals in the connectional or functional organization of prefrontal cortex (e.g., Eichenbaum et al., ’83; Kolb, ’84; Kolb and Whishaw, ’85). It is noteworthy that the contending points of view regarding comparative frontal lobe anatomy were established prior to advent of many of the techniques now commonly employed in neuroanatomy, such as the horseradish peroxidase and autoradiographic methods for tracing connections. These and other technical innovations have greatly altered our understanding of primate GFC organization. At midcentury, primate GFC was regarded as a relatively homogeneous entity, envincing little regional architectonic or functional specialization (e.g., Bonin, ’48; Bonin and Bailey, ’47; Lashley and Clark, ’46).Evidence ‘The origin of the infelicitous term “prefrontal,” as applied to the cortex, has long been something of a mystery. Divac (’88) attributes its first use to Richard Owen, a contemporary of Darwin and one of the leading authorities on comparative vertebrate anatomy in his day (Russell, ’16). This suggests a derivation for “prefrontal cortex.” The lobes of the mammalian cerebrum are named for the bones which overlay them: occipital, parietal, temporal, and frontal. There is no prefrontal bone in modern mammals. There is a prefrontal bone in other vertebrates, however, as Owen no doubt appreciated. This bone lies rostral t o the frontal bone, occupying the anterior wall of the bony orbit in many vertebrate groups (see Romer, ’70, or similar texts). Thus, the prefrontal bone bears an anatomical relationship to the frontal bone comparable to that between the “silent” prefrontal cortex and the electricallyexcitable frontal h e . , precentral) cortex.
LUS MD MF MFc MFr MOS MPOS Oa-p OB OF0 OLS Orb
0s
OTrS OTS OTu Para PCS PFC Pir PL POS PrCO PS Put RcS RoS RS RSa RSP SPS STS
lunate sulcus mediodorsal thalamic nucleus medial frontal areas caudal division of medial frontal cortex rostral division of medial frontal cortex medial orbital sulcus medial parieto-occipital sulcus periallocortical division of agranular orbital cortex olfactory bulb orbitofrontal opercular area olfactory sulcus orbital cortex orbital sulcus transverse occipital sulcus occipito-temporalsulcus olfactory tubercle paralimbic cortex paracalcarine sulcus prefrontal cortex piriform cortex prelimbic area parieto-occipital sulcus precentral opercular area principal sulcus putamen retrocalcarine sulcus rostral sulcus rhinal sulcus rhinal sulcus, anterior part rhinal sulcus, posterior part splenial sulcus superior temporal sulcus
from modern studies, however, and in particular from connectional studies of macaque monkeys (e.g., Cavada and Goldman-Rakic, ’89; Preuss and Goldman-Rakic, ’89; Seltzer and Pandya, ’89) and studies of metabolic activity in the human brain (Roland, ’84; Roland et al., ’87), suggests a much greater degree of regional structural and functional parcellation in the GFC than was suspected by most earlier workers. This new perspective on GFC organization raises old issues concerning its evolution: Do all mammals possess the full complement of GFC areas present in large-brained anthropoid primates, such as macaques and humans? If not, which areas are present only in anthropoids and which are more widely distributed among mammals? In order to address evolutionary issues such as these, we have undertaken a comparative anatomical study of primate granular frontal cortex. Specifically, we have employed modern architectonic and connectional techniques to compare the structure of GFC in two kinds of primates: macaque monkeys (genus Macaca), a group of anthropoid primates closely related to humans, and galagos (also known as bushbabies; genus Galago),which are representatives of the strepsirhine (or “prosimian”) primates. Some comments about strepsirhine primates, and why they are appropriate as a comparison group for anthropoids, are in order. As illustrated in Figure 1, the order Primates has two major branches, the haplorhines and strepsirhines. The haplorhine branch consists mainly of the familiar anthropoid primates-that is, monkeys, apes, and humans. The strepsirhines consist of lemurs, lorises, and galagos. In comparative anatomy, the evolutionary specializations of one group of animals are assessed by comparing them to closely related groups. So, strepsirhines such as Galago have long been used as an “outgroup” in the study of anthropoid specializations, as well as to determine what biological characteristics are common to all primates. This strategy has been pursued both in biological anthropology
ARCHITECTURE OF PRIMATE GRANULAR FRONTAL CORTEX
~-~PEalBidaalB Lemurs
Lorises
Galagos
431
-4 Tarsiers
New World Monkeys
Apes & Humans
Old World Monkeys
-
Primates Fig. 1. Evolutionary relationships of the placental mammals and of the primates. The relationships of the major groups of Iiving primates, as shown above, are widely agreed upon by modern workers (see, e.g., Fleagle, '88; Martin, '90; Richard, '85; Szalay and Delson, '79). Primates are composed of two major groups, strepsirhines and hap-
lorhines. The living stepsirhines are a fairly conservative group both anatomically and behaviorally. The haplorhines consist of the tarsiers (a single genus of small, nocturnal animals) and the familiar anthropoid ("higher") primates.
(reviewed by Cartmill, '82) and in neurobiology (e.g., Allman, '82; Carlson, '85; Diamond et al., '85; Kaas, '83). Strepsirhines are particularly well suited for this comparative role: although they are not literally primitive primates (they are in fact a diverse group with their own set of specializations), they are nonetheless evolutionarily conservative. That is, they retain many anatomical features of the earliest primates (and of related non-primate mammals) which have been modified during the evolution of anthropoid primates (Fleagle, '88; Martin, '90). Furthermore, there is good reason to suppose that neural organization is conservative in strepsirhines relative to anthropoids. Strepsirhine brains are small, averaging about half the size of anthropoid brains for a given body weight (Jerison, '73). Also, electrophysiological mapping studies indicate that while strepsirhines possess many of the visual and somatosensory cortical areas which are found in anthropoids, anthropoids have additional representations not found in strepsirhines (Allman et al., '79; Kaas, '83). Based on these considerations, we thought it likely that by comparing the strepsirhine Galago and the anthropoid Macaca, we could identify anthropoid specializations of GFC organization. In addition, by identifying GFC characteristics which are shared by both major groups of primates (strepsirhines and anthropoids), we can better characterize the basic plan of primate frontal lobe organization, and so provide a firmer basis for comparing primates to other mammals. In this paper, we report the results of a comparative myelo- and cytoarchitectonic study of granular frontal cortex of Galago and Macaca. The specific goals of this study were to make an areal parcellation of the GFC in
Galago and Macaca, and an initial interpretation of the homologies among areas, based on similarities of architectonic structure and location within the cortical mantle. This interpretation will subsequently be tested with connectional information (Preuss and Goldman-Rakic, '91b). There have been many previous architectonic studies of primate GFC, in macaques and other anthropoids (e.g., Barbas and Pandya, '89; Bonin and Bailey, '47; Brodmann, '05, '09; Rosabel, '67; Walker, '40) as well as in galagos and related strepsirhines (Bonin, '45; Brodmann, '08, '09, '12; Le Gros Clark, '31; Sanides and Krishnamurti, '67; Zilles and Schleicher, '80; Zilles et al., '79a,b, '82). (The widely used cytoarchitectonic maps of Brodmann "091 and Walker "401 are shown in Fig. 2.) In our view, these studies do not provide an adequate comparative foundation for interpreting the evolution of primate GFC. Their major shortcoming is a lack of agreement regarding the number and location of areal subdivisions within the GFC, even for a single species like Macaca mulatta (compare, for example, Bonin and Bailey, '47;and Walker, '40).There are even disagreements regarding the precise extent and borders of the granular frontal cortex, in part because there is no sharp discontinuity between cortex which possesses an internal granular layer (IV)and cortex which lacks it. For instance, Roberts and Akert ('63) and Barbas and Pandya ('87) would include much of the frontal opercular cortex within the granular frontal region, because it has a layer IV1although the same cortex was included within the agranular, precentral region by Brodmann ('09) and Walker ('40). Clearly, in order to compare the areal organization of GFC of Galago and Macaca, it is necessary to delimit the GFC in a consistent manner in both taxa and to have an
T.M. PREUSS AND P.S. GOLDMAN-RAKIC
432
12
26
Cercopithecus
i0
