AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 89:299-308 (1992)
Meningeal Arteries in Rhesus Macaques (Macaca rnulatta): Implications for Vascular Evolution in Anthropoids DEAN FALK AND PHILIP NICHOLLS Department of Anthropology, State University of New York at Albany, Albany, New York 12222
Middle meningeal artery, Rhesus, Endocast, CraKEY WORDS nial vasculature, Anterior meningeal artery, Cranial capacity
ABSTRACT The branching patterns of meningeal arteries are reported for 200 endocast hemispheres representing rhesus monkeys (Macaca mulatta) of known cranial capacity. We detect five basic patterns for the branching of the anterior division of the middle meningeal artery and its relationship with the anterior meningeal artery. These results confirm and elaborate trends published for much smaller samples that were based on direct dissections of rhesus monkey arterial patterns. The most common pattern is that in which the anterior meningeal artery dominates the blood supply above the rostra1 part of the middle cranial fossa. Analysis of cranial capacities reveals that presence of this pattern on both sides of endocasts is associated with increased cranial capacity. When studied in light of published reports of anatomical dissections of cranial arteries in apes and human embryological data, the anterior meningeal artery in rhesus monkeys appears to be a possible homologue of the lacrimal meningeal artery in apes and the anterior branch of the middle meningeal artery in humans. This finding provides a step towards understanding cranial vasculature homologies that may be useful for accurately scoring the branching patterns of the meningeal arteries in monkeys, apes, and humans. o 1992 Wiley-Liss, Inc. The human middle meningeal artery is a close proximity, it has been suggested that branch of the maxillary artery, which stems meningeal arteries and veins may communifrom the external carotid. It supplies large cate without the intervention of a capillary areas of the calvarium and the dura mater network; Jones, 1912.) that covers the human brain (Batson, 1944; Jones’ opinion is supported by dissections Crosby et al., 1962). The veins and sinuses of humans, which reveal that the grooves (e.g., petrosquamous and lateral extension appear to contain two veins, the paraof the sphenoparietal) that accompany the meningeal veins, lying on either side of their middle meningeal artery drain the diploe, respective arteries (Crosby et al., 1962). The dura mater, and periosteum of the skull double appearance of the veins outlining the (Dahl and Edvinsson, 1988). Some authors artery in dissection material has been hyclaim that the contents of the meningeal pothesized to be the result of compression of grooves in skulls are arteries, while others the vein between the artery and bone during claim they are veins. According to Jones (19121, although the margins of most of the grooves are actually caused by veins, these Received October 4,1991; accepted May 19,1992. grooves nevertheless also reflect the courses Address correspondence to Dean Falk, Department of Anthroof smaller arteries that usually travel on the pology, State University of New York at Albany, Albany, NY surfaces of veins. (In fact, because of their 12222. 0 1992 WILEY-LISS, INC
300
D. FALK AND P. NICHOLIS
development (Padget, 1956). (Padget’s hypothesis is based on Jones’ (1912) earlier observations of variation in arterialhenous configurations in adults and limited fetal material.) Jones notes that “strangely enough a great deal that is not evident from a study of the grooves themselves is made conspicuous” by endocasts (1912:233). He also correctly observes that endocasts frequently reveal “minute, and often zigzag beading marks” which represent the paths of the narrower meningeal arteries that course on or near the wider veins. Jones adds that in most endocasts “it is easy to separate arterial from venous impressions” (1912:234), although insulation or buffering of meningeal arteries by their larger companion veins sometimes prevents arteries from leaving distinct imprints within the wider venous grooves. Since endocasts can be prepared from readily available skulls in museum collections, they afford an opportunity to study cranial vasculature in larger samples than would be available from scarce cadaveric resources, When studied in light of anatomical dissections of cranial vasculature in monkeys and apes, as well as relevant embryological data for humans, endocasts provide an excellent tool for exploring the evolution of meningeal vessels in higher primates. Because meningeal vessels are represented by clear grooves on the insides of the anterior and middle cranial fossae, they have been widely studied in human skulls and fossilized hominid cranial remains. Some students of meningeal vessels have speculated about vascular evolution in fossil hominids (Weidenreich, 1938) and proposed that meningeal vascular patterns can be used to sort them into different groups (Saban, 1983, 1985, 1986). However, these studies lack adequate comparative data regarding meningeal patterns in monkeys and apes. According to Diamond (19881, rhesus macaques (Macaca mulatta) retain a relatively conservative endocranial vascular configuration, and therefore provide a reasonable comparative model for the vascular state from which ape and human arterial patterns are derived. The purpose of the present paper is to present new information about meningeal arteries determined from a
large sample of rhesus monkey endocasts, and to discuss the implications of these data for understanding the evolution of cranial vasculature in apes and humans. MATERIALS AND METHODS Meningeal arterial patterns were studied from endocasts that were prepared by D.F. from skulls of rhesus monkeys (Macaca mulatta), using standard techniques (Falk, 1978).All observations were scored independently by both authors, and only specimens for which there was agreement were included in our sample of 200 hemispheres. In most cases, both hemispheres of an endocast were scored. Specimens represented both sexes and spanned from juveniles to adults. The relationship between meningeal arterial pattern and brain size was investigated, using cranial capacities collected by D.F. (accordingto standard methods; Falk, 1976) in conjunction with endocast preparation. Finally, referring to studies of direct dissections of pongid cranial vascular anatomy and human embryological data (Muller, 1977), we explored the implications of our findings for understanding the homologies between branches of the anterior meningeal arteries in human and nonhuman primates. The middle meningeal artery frequently enters the skull of rhesus monkeys through the lateral end of the foramen ovale (Castelli and Huelke, 1965a). However, Diamond (1988:57) observes two dissected specimens ofMacaca mulafta in which the artery “penetrates the cranial base by passing through the petrosphenoid fissure, posterior and lateral to the foramen ovale,” and reports that this condition exists in many cercopithecoids. Once the middle meningeal artery enters the skull, it divides into anterior (frontal) and posterior (parietal) branches. Although the subsequent course of the posterior branch is relatively constant in rhesus monkeys (Castelli and Huelke, 1965a1, the path taken by the anterior branch of the middle meningeal artery exhibits various patterns, which we scored from our endocasts. Diamond has adopted a system of nomenclature for the meningeal arteries that is based on stapedial artery homologies determined from embryological studies as well as comparison of a wide spectrum of primates
MENINGEAL ARTERLIES IN MACAQUES
and selected species from other orders (1988; 1991a,b). Diamonds system therefore differs from conventional nomenclature in that arteries are labeled according to their developmentaVphylogenetic histories rather than their origin, distribution, and function in postnatal individuals and extant species. Thus, if Diamond’s system were applied to macaques, the anterior meningeal artery that originates (via other branches) from the internal carotid artery would have the same name (“meningeal branch of the anterior division of the ramus superior of the stapedial artery”) as the anterior branch of the middle meningeal artery that originates from the maxillary artery of the external carotid artery, because he believes both of these arteries developed from the same branch of the stapedial artery (see, however, Muller, 1977). Although Diamond’s system would be appropriate for studies that focus strictly on homologies, it is inappropriate for the purpose of the present study, which is to identify and establish the range of variation in postnatal branching patterns of the meningeal arteries of one extant species of higher primate. Moreover, adoption of Diamond’s nomenclature would also result in a confusing and cumbersome terminology that is incompatible with that used in the wider literature on primate vascular anatomy. This is not to say that we discount the importance of embryological studies or that we think vascular homologies are unimportant. Indeed, we use these concepts to assess the implications of the findings for rhesus monkeys for vascular evolution in apes and humans. However, for reasons given above, we do not advocate changing the names of vascular structures to reflect their embryological or evolutionary histories. The scoring procedures used in this study were based on anatomical studies of cranial vasculature in macaques. These studies from the literature relied on direct dissection of cadavers, dissection of latex-injected specimens, corrosion preparations, and cleared specimens (Castelli and Huelke, 1965a,b; Weinstein and Hedges, 1962). Illustrations of three main patterns of distribution for the anterior branch of the middle meningeal artery, determined from 20 specimens by Castelli and Huelke (1965a), were
301
used to score meningeal artery patterns from rhesus endocasts. In the first pattern, the anterior division of the middle meningeal artery is reduced to a small anastomotic connection, which joins the anterior meningeal artery (which is off of the ophthalmic branch of the internal carotid artery) that supplies the side of the skull above the middle cranial fossa (see Fig. 2A). In the second pattern, the anterior meningeal artery and the anterior branch of the middle meningeal artery both provide branches to the side of the skull above the middle cranial fossa. These branches may be separate, or united t o varying degrees (Fig. 2B-D). Finally, in Castelli and Huelke’s third pattern, the middle meningeal artery supplies most of the sides of the anterior and middle cranial fossae while the anterior meningeal artery appears to be very small or absent (Fig. 2E). Below, we report the frequencies of these three patterns and two additional variations in a large sample of rhesus endocasts. We also explore the association between these patterns and cranial capacity (brain size). Because the origin of the middle branch of the middle meningeal artery has been a point of focus in the literature on human meningeal patterns, we scored that trait in the subset of rhesus endocasts that had the potential for having identifiable middle branches, namely those with patterns represented by Figure 2C,E. (The remaining patterns did not lend theniselves to such scoring because they lacked unambiguous extensions of a separate anterior branch of the middle meningeal artery onto the frontal and parietal surfaces.) Thus, using the system developed for humans by Adachi (1928), a hemisphere was scored as 0 if its middle meningeal artery lacked a middle branch; type I if a middle branch stemmed from the anterior branch of the middle meningeal vessels; type I1 if the middle branch was off of the posterior branch; and type I11 if clear middle branches stemmed from both anterior and posterior branches of the middle meningeal vessels (Fig. 1). RESULTS We observed five basic patterns for branching of the anterior division of the middle meningeal artery and its relationship to the anterior meningeal artery in the
302
D. FALK AND P. NICHOLLS
Fig. 1. Branching patterns of the middle meningeal artery that are frequently used to score humans, right sides. Type I, middle branch (Mmj arises from the anterior branch (Ma); Type 11, middle branch arises from the posterior branch (Mp); Type 111, middle branches arise from both anterior and posterior branches of the middle meningeal artery. Classification after Adachi (1928); illustration modified from von Bonin (1963).
200 rhesus monkey hemispheres (Fig. 2AE). The frequencies for each pattern are presented separately for right and left hemispheres in Table 1. The most frequent pattern (Fig. 2A) occurred in 69% of the endocast hemispheres. In this pattern, the anterior (or frontal) branch of the middle meningeal artery is truncated or reduced to a small connection at the junction with the anterior meningeal artery that supplies the side of the skull and dura above the middle cranial fossa (Castelli and Huelke, 1965a). Patterns R, C, and D in Figure 2 are all variations of the second most common pattern reported from dissection material (Castelli and Huelke, 1965a1, in which the anterior meningeal artery and the anterior branch of the middle meningeal artery both supply the side of the skull above the middle cranial fossa. That area is supplied by a common stem that is formed by the two arteries in pattern B, which is very rare in our sample. In pattern C, the two arteries are separated along their entire lengths, with the anterior branch of the middle meningeal running more or less parallel with and caudal to the anterior meningeal. (In one interesting variant of this pattern, the anterior meningeal artery crossed over the anterior branch of the middle meningeal.) Instead of running parallel, the two branches in pattern D join near their proxi-
mal ends, forming a triangle. Patterns B, C, and D together form 24.5% of our endocast sample. Finally, in pattern E (6.5% of the sample), the middle meningeal artery supplies most of the sides of the anterior and middle cranial fossae while the anterior meningeal artery appears to be absent. The posterior or parietal branch of the middle meningeal artery supplies the posterior part of the middle cranial fossa (frequently traveling with the petrosquamous sinus) and then sends a branch onto the posterior cranial vault and dura above the petrous portion of the temporal bone. We confirmed Castelli and Huelke’s (1965a) report that both the position and occurrence of this artery are relatively constant. However, we also observed that the extent of branching on the cranial vault varies from extremely simple t o elaborate, depending on the extent and elaboration of the posterior meningeal artery, which enters the skull near the junction of the transverse and sigrnoid sinuses. If the posterior meningeal artery is elaborate and extends rostrally across the parietal bone, the posterior branch of the middle meningeal artery tends to be modest, and vice versa (Fig. 3). Interestingly, Diamond (1988) observed a similar relationship for these two arteries in platyrrhine monkeys. We applied Adachi’s (1928) system for scoring branching patterns of the middle
MENINGEAL ARTERIES IN MACAQUES
D
303
E
Fig. 2. The five basic branching patterns of the anterior division of the middle meningeal artery and its relationship to the anterior meningeal artery in rhesus monkey endocasts, right sides. A, anterior meningeal artery; P, posterior meningeal artery; M, middle meningeal artery; Ma, anterior branch of the middle meningeal artery; Mp, posterior branch of the middle meningeal artery. The most frequent pattern observed was A, which appeared in 69%of 200 hemispheres. See text for other details.
meningeal artery in humans (described above) to the branching patterns of the 13 hemispheres that exhibited pattern E and to 38 of the 41 hemispheres with pattern C . For
the latter pattern, a middle branch originated off of the anterior branch of the middle meningeal artery (type I) in 8 specimens, off of the posterior branch (type 11) in 8 speci-
304
D. FALK AND P. NICHOLLS
anterior middle cranial fossa are supplied only by the anterior meningeal artery (patrhesus monkev endocasts tern A on both sides) than in specimens Right side Total Left side where both sides are supplied solely or parn n n (961 Pattern tially by the anterior branch of the middle 75 138 (69.0) A 63 B 01 02 03 (01.5) meningeal pattern (i.e., the group of en18 41 (20.5) docasts whose two sides exhibit any combiC 23 1) 04 01 05 (02.5) nation of B, C , D, and El. In the former 04 13 (06.5) E 09 group, cranial capacity averages 100.4 cm3 Total 100 100 200 (100) (range = 85-130 cm'; n = 471, whereas the mean for the latter group is 92.8 cm3 (range = 70-113 cm3; n = 12). This differmen&, f i ~ mboth branches (type III) in 8 ence in cranial capacity is statistically sigspecimens,andfromneitherbranchin 14hemi- nificant, t(57) = 2.20, P = .031, two-tailed. spheres. The ratings for pattern E were much Thus, there is a positive relationship besimpler. In all 13 hemispheres, the middle tween brain size and increased reliance on branch originated from the anterior branch the anterior meningeal and therefore internal carotid artery (i.e., when A appears on of the middle meningeal artery (type I). To determine if meningeal arterial pat- both sides) for supplying the dura and craterns are differentially represented in the nial vault above the middle cranial fossa. two sexes, we performed a chi-square test on Because the mean ages of the two groups endocasts drawn from our entire sample does not differ significantly (6.87 years for (with three categories: A, B C + D, and E). the first group consisting of pattern A on The result was x2 = 4.96, df = 2, which was both sides, 7.04 years for the group of ennot statistically significant. Thus, there are docasts with any combination of B, C , D, and no statistically significant sex differences in E, t < 11, the significant relationship bethe manifestation of meningeal arterial pat- tween meningeal pattern and cranial capacterns in the macaques we sampled. We also ity does not appear to be due to age. performed a similar chi-square test to determine if the different patterns were differenDISCUSSION AND CONCLUSIONS tially represented in the left and right hemiAlthough our observations of endocasts spheres across our entire sample. The result was x2 = 3.97, df = 2, which was not statis- from rhesus monkeys confirmed reports tically significant. A third chi-square test that meningeal arteries and veins frewas performed on two categories, A versus quently travel together, our investigation foB + C + D (some of the cells for E were cused on the arteries because of the availempty), to determine if there is a depen- ability of their descriptions from direct dence between patterns in the left and right dissections of Macaca mulatta (Castelli and hemispherees within those 73 individuals Huelke, 1965a). Indeed, we would not have for whom these meningeal patterns were been able to accurately interpret meningeal available from both hemispheres. The result arterial patterns from rhesus endocasts was x2 = 2.12, df = 1,which was not statis- without such reports. For example, had we tically significant. We therefore conclude relied on the literature for human that meningeal patterns are not differen- meningeal patterns, we would have misintially represented in left and right hemi- terpreted the most frequent rhesus monkey spheres and that, within individual speci- pattern (A) by erroneously assuming that mens, the pattern of one hemisphere does the anterior meningeal artery was the antenot predict the pattern of the other. (Inter- rior branch of the middle meningeal artery. estingly, Diamond (1988) reached another (This is because the location and visual configuration with respect to the anterior diviconclusion for hominoids.) We found that cranial capacity is signifi- sion of the middle meningeal artery superficantly larger in those endocasts in which the cially resembles the human condition.) The dura and both sides of the skull above the importance of interpreting meningeal patTABLE I , Frequencies of the fiue basic patterns
of the middle meningeal artery determined from
+
MENINGEAL ARTERIES IN MACAQUES
305
tends to be invcrse!g rclatcd t o Fig 3 The extent of branching ofthe posterior memnge?! srtery the amount of branching of the posterior branch of the middle meningeal artery (Mp), as illustrated by left and right posterior posterolateral views taken from two different rhesus monkey endocasts Lambdoid sutures and midlines are represented by dashes See text for discussion
terns on endocasts in light of direct dissections of appropriate species cannot be overstressed. Because of our large sample size, we were able to confirm and elaborate the trends and patterns observed by Castelli and Huelke (1965a) from their dissections of 40 right and left sides combined. Castelli and Huelke’s most frequent pattern occurred in 52% of the sides and is equivalent to the most frequent pattern (A) that occurred in our sample (69% of 200 hemispheres). Our study extends the findings of Castelli and Huelke by reporting frequencies for the individual variations of their second most frequent pattern (28% of their sample). Thus, patterns B, C , and D are all variations of this pattern and together total 24.5% of our sample. Finally, our pattern E (6.5%) was the least frequent of the three patterns that Castelli and Huelke described (20%). As the above shows, the overall distribution of the frequencies of the three meningeal patterns that we observed in our endocasts (n = 138, 49, and 13 for Castelli and Huelke’s Patterns 1, 2 and 3, respectively, yielding proportions of .69, .245, and .065) is similar to the overall distribution that Castelli and Huelke observed in direct dissections of macaque cerebral arteries (n = 21, 11, and 8 for Patterns 1, 2 and 3, respectively, yielding proportions of 5 2 , .28, and .20). Nevertheless, the patterns of our and Castelli and Huelke’s overall frequency distributions were statistically different
(x2= 8.43, df = 2, P < .02). It appears that this overall significant difference was largely due to our observing a greater proportion of Pattern 1(or A) but a smaller proportion of Pattern 3 (or E) than Castelli and Huelke observed. It is possible that some endocasts that actually represented pattern E were misidentified as belonging to pattern A. This could occur if the anterior meningeal vessels that normally appear as pattern E on an endocast had been located somewhat rostrally and were buried within the anterior end of the Sylvian fissure in the living monkey. Despite this possible “noise”in our data, we are satisfied that our overall results reveal the trends predicted from the dissection data, and provide additional details about meningeal arterial patterns in rhesus monkeys. We also note that, although their frequency may be underestimated for the above reasons, the set of endocasts classified as manifesting pattern E does not appear to contain specimens that have been misidentified. As shown in the results section, our exercise in applying Adachi’s method for classifying the origin of the middle meningeal artery to rhesus monkey endocasts did not work very well for pattern C, since 14 of 38 hemispheres did not even appear to have a middle branch and the other 24 were evenly distributed across the other three possible conditions. On the other hand, all 13 of the hemispheres manifesting pattern E could be scored. A middle branch originated off of the
306
D. FALK AND P. NICHOLLS
anterior branch of the middle meningeal artery in all of them. We can therefore conclude that in rhesus monkeys where there is a predominance of the middle meningeal artery over the anterior meningeal artery (i.e., on the side of the skull above the middle cranial fossa), this pattern is associated with branching of the anterior division of the middle meningeal artery. It would be premature to compare these data for rhesus monkeys with those for humans, however, without first establishing the ~7ariouspatterns and their frequencies for meningeal arteries in the great apes. The fact that the group of endocasts that relied solely on the anterior meningeal artery to supply the side of the skull and dura mater above the anterior part of the middle cranial fossa (pattern A on both sides) have a significantly larger mean cranial capacity than the group o f remaining endocasts shows that increased brain size in rhesus monkeys is associated with increased dependence on the internal carotid artery (from whence stems the ophthalmic artery that gives rise to the anterior meningeal artery). It would be interesting to determine whether our findings based on intraspecific variation in rhesus monkeys bear on a phylogenetic analysis. In other words, do the great apes (who are bigger brained than rhesus monkeys) rely on branches of the ophthalmic artery rather than the middle meningeal artery lo supply the sides of the skull above the middle cranial fossae? Direct dissections o f cranial arteries in one gorilla and one orangutan in conjunction with observations of grooves of meningeal arteries in skulls of two additional gorillas suggest that the answer to this question is yes (Muller, 1977). In these pongids, the most rostral portion of the sides of the skull above the middle cranial fossae are supplied by branches from the meningeal lacrimal artery, which is a branch of the ophthalmic artery that stems from the internal carotid artery. (Muller observes that these extra-orbital branches of the lacrimal artery enter the braincase through the supra-orbital fissure or, alternatively, through the cranio-orbital foramen.) The middle meningeal vessels are distinct from and caudal to these branches. Although Dia-
mond (1991a) disagrees with Muller’s conclusions regarding homologies, his own dissection of one chimpanzee specimen revealed that, on both sides, the meningeal branch to the frontoparietal area was “fed in retrograde fashion by the ophthalmicthalmic artery” (p. 235). Furthermore, Diamond notes that on the right side of one of two orangutan specimens that he dissected, “the definitive lacrimal artery gave off a recurrent branch that entered the middle cranial fossn through 3 foramer, . . .” fp. 23;;. (Although discussion of the controversies surrounding meningeal-orbital arteries is beyond the scope of this paper, this topic is reviewed by Diamond, 1991a.) The pattern reported by Muller (1977) is similar to the most common pattern reported from rhesus monkey dissections, in which the anterior meningeal artery occupies the same rostral location and also stems from the ophthalmic artery. (Compare Fig. 1 of Castelli and Huelke, 1965a with Fig. 5 of Muller, 1977; see Fig. 4.) Thus the meningeal lacrimal artery of pongids and the anterior meningeal artery of rhesus monkeys appear to be likely homologues. Muller (1977) proposes that the anterior branch of the middle meningeal artery in humans is the homologue of the pongid meningeal lacrimal artery, based on a study of cranial vascular development in a series of injected human fetuses that varied from 143 to 290 mm crown-rump length. During fetal development, the lacrimal artery (which is a branch of the ophthalmic) enters the middle cranial fossa from the orbit where it connects with the middle meningeal artery medially, while delivering a named branch (rml, the principal meningeal branch of the lacrimal artery) onto the wall of the cranium laterally. Rml ascends the vault between the frontal and parietal bones, and (as development proceeds) the middle meningeal artery eventually takes it over (“annexes”it) through the connection mentioned above. Thus, the anterior branch of the middle meningeal artery of humans appears homologous not only with the meningeal lacrimal artery of pongids (Muller, 1977), but also with the anterior meningeal artery of rhesus monkeys. (For an alternative view, see Dia-
MENINGEAL ARTERJES IN MACAQUES
307
ference is that the homologous artery is usually distinct from the middle meningeal artery in monkeys and apes, but has become incorporated as the anterior branch of the middle meningeal artery in humans. Muller’s comparative study of apes and humans therefore led to the conclusion that the “middle meningeal artery gains in importance as mammalian evolution proceeds” (1977:218).However, it remains to be seen if and how the patterns described above for rhesus monkeys are distributed within and across samples of great apes, or if apes manifest additional arterial variations. We hope to pursue these issues in future studies and to apply our findings to furthering the understanding of the evolution of cranial vasculature in monkeys, apes, and humans. ACKNOWLEDGMENTS We thank Marc Schmidt for help with illustrations and are grateful to Jim Neely and Michael Diamond for comments on the manuscript. The University of Puerto Rico is acknowledged for providing free access to the Cayo Santiago skeletal collection. This research was supported by NSF grant No. BNS-9008179.
I Fig. 4. Most frequent arterial branching pattern observed in rhesus monkey endocasts (above) compared to that of a 3 year old gorilla (middle) and an adult gorilla (below), right sides. The gorilla patterns were determined by Muller (1977). In apes, the most rostra1 portion of the sides of the skull above the middle cranial fossae are supplied by branches of the meningeal lacrimal artery (La), which is a branch of the ophthalmic artery. La appears to be homologous with the anterior meningeal artery (A) of rhesus monkeys because of its location, origin, and configuration with respect to the anterior branch of the middle meningeal artery. See text for discussion and Figure 2 for other abbreviations.
mond, 1991a.) All three of the proposed homologues derive (developmentally) from the ophthalmic artery, and supply the same regions of the skull and dura mater. The dif-
LITERATURE CITED
Adachi B (1928)Das Arteriensystem der Japaner. Band 1. Kyoto: Verlag der Kaiserlich-Japanischen Universitat zu Kyoto. Batson OV !1944) Anatomical problems concerned in the study of cerebral blood flow. Fed. Proc. 3r139-144. Bonin G von (1963) The Evolution of the Human Brain. Chicago: The University of Chicago Press. Castelli WA, and Huelke DF (1965a)The arterial supply of the dura mater of the rhesus monkey. Anat. Rec. 152155-160. Castelli WA, and Huelke DF (1965b) The arterial system of the head and neck of the rhesus monkey with emphasis on the external carotid system. Am. J. Anat. I 1 6;149-170. Crosby EC, Humphrey T. and Lauer EW (1962) Correlative Anatomy of the Nervous System. New York: Macmillan. Dahl E, and Edvinsson L (1988) Anatomical organization of cerebral and extracerebral vasculature. In J Olesen and L Edvinsson (eds.): Basic Mechanisms of Headache. Amsterdam, Elsevier Sci. Pub., pp. 2 7 4 7 . Diamond MK (1988) Cephalic Vascular Evolution and Development in Primates: The Stapedial Artery and its Companion Venous Sinuses. PhD dissertation, University of Chicago. Ann Arbor: University Microfilms.
308
D. FALK AND P. NICHOLLS
Diamond MK (1991a1 Homologies of the meningeal-orbital arteries of humans: A reappraisal. J . Anat. 17#:223-241. Diamond MK (1991b)Homologies of the stapedial artery in humans, with a reconstruction of the primitive stapedial artery configuration of euprimates. Am. J. Phys. Anthropol. #4:433-462. Falk D (1976)External Neuroanatomy of the Cercopithecoidea. PhD dissertation, University of Michigan. Ann Arbor: University Microfilms. Falk D (19781 External Neuroanatomy of Old World Monkeys (Cei-copithecoidea).Contrib. Primatol. 15t195. Jones FW (1912) On the grooves upon the ossa parietalin rommonly said to he raiised hy the arteria moningea media. J. Anat. 46:22&238. Muller F (1977) The development of the anterior falcate and lacrimal arteries in the human. Anat. Embryol. 150;207-227.
Padget DH (1956) The cranial venous system in man in reference to development, adult configuration, and relation to the arteries. Am. J. Anat. 98t307-355. Saban R (1983) Les veines meningees moyennes des australopitheques. Bull. Mem. SOC.d’Anthrop. (Paris) 1Ot313-324. Saban R (1985) Identification des stades evolutifs des hommes fossiles d’apres les veines meningees. Actual. Odonto-Stomatol. 152t693-712. Saban R (1986) Veines meningees et hominisation. Anthropos (Brno)23r15-33. Weidenreich F (1938) The ramification of the middle meningeal artery in fossil hominids and its bearing upon phylogenetic problems. Palaeontologia Sinica New Sw D.?:l-l6 Weinstein JD, and Hedges TR (1962) Studies of intracranial and orbital vasculature of the rhesus monkey (Macaca mulatta).Anat. Rec. 144r37-41.
Meningeal Arteries in Rhesus Macaques (Macaca rnulatta): Implications for Vascular Evolution in Anthropoids DEAN FALK AND PHILIP NICHOLLS Department of Anthropology, State University of New York at Albany, Albany, New York 12222
Middle meningeal artery, Rhesus, Endocast, CraKEY WORDS nial vasculature, Anterior meningeal artery, Cranial capacity
ABSTRACT The branching patterns of meningeal arteries are reported for 200 endocast hemispheres representing rhesus monkeys (Macaca mulatta) of known cranial capacity. We detect five basic patterns for the branching of the anterior division of the middle meningeal artery and its relationship with the anterior meningeal artery. These results confirm and elaborate trends published for much smaller samples that were based on direct dissections of rhesus monkey arterial patterns. The most common pattern is that in which the anterior meningeal artery dominates the blood supply above the rostra1 part of the middle cranial fossa. Analysis of cranial capacities reveals that presence of this pattern on both sides of endocasts is associated with increased cranial capacity. When studied in light of published reports of anatomical dissections of cranial arteries in apes and human embryological data, the anterior meningeal artery in rhesus monkeys appears to be a possible homologue of the lacrimal meningeal artery in apes and the anterior branch of the middle meningeal artery in humans. This finding provides a step towards understanding cranial vasculature homologies that may be useful for accurately scoring the branching patterns of the meningeal arteries in monkeys, apes, and humans. o 1992 Wiley-Liss, Inc. The human middle meningeal artery is a close proximity, it has been suggested that branch of the maxillary artery, which stems meningeal arteries and veins may communifrom the external carotid. It supplies large cate without the intervention of a capillary areas of the calvarium and the dura mater network; Jones, 1912.) that covers the human brain (Batson, 1944; Jones’ opinion is supported by dissections Crosby et al., 1962). The veins and sinuses of humans, which reveal that the grooves (e.g., petrosquamous and lateral extension appear to contain two veins, the paraof the sphenoparietal) that accompany the meningeal veins, lying on either side of their middle meningeal artery drain the diploe, respective arteries (Crosby et al., 1962). The dura mater, and periosteum of the skull double appearance of the veins outlining the (Dahl and Edvinsson, 1988). Some authors artery in dissection material has been hyclaim that the contents of the meningeal pothesized to be the result of compression of grooves in skulls are arteries, while others the vein between the artery and bone during claim they are veins. According to Jones (19121, although the margins of most of the grooves are actually caused by veins, these Received October 4,1991; accepted May 19,1992. grooves nevertheless also reflect the courses Address correspondence to Dean Falk, Department of Anthroof smaller arteries that usually travel on the pology, State University of New York at Albany, Albany, NY surfaces of veins. (In fact, because of their 12222. 0 1992 WILEY-LISS, INC
300
D. FALK AND P. NICHOLIS
development (Padget, 1956). (Padget’s hypothesis is based on Jones’ (1912) earlier observations of variation in arterialhenous configurations in adults and limited fetal material.) Jones notes that “strangely enough a great deal that is not evident from a study of the grooves themselves is made conspicuous” by endocasts (1912:233). He also correctly observes that endocasts frequently reveal “minute, and often zigzag beading marks” which represent the paths of the narrower meningeal arteries that course on or near the wider veins. Jones adds that in most endocasts “it is easy to separate arterial from venous impressions” (1912:234), although insulation or buffering of meningeal arteries by their larger companion veins sometimes prevents arteries from leaving distinct imprints within the wider venous grooves. Since endocasts can be prepared from readily available skulls in museum collections, they afford an opportunity to study cranial vasculature in larger samples than would be available from scarce cadaveric resources, When studied in light of anatomical dissections of cranial vasculature in monkeys and apes, as well as relevant embryological data for humans, endocasts provide an excellent tool for exploring the evolution of meningeal vessels in higher primates. Because meningeal vessels are represented by clear grooves on the insides of the anterior and middle cranial fossae, they have been widely studied in human skulls and fossilized hominid cranial remains. Some students of meningeal vessels have speculated about vascular evolution in fossil hominids (Weidenreich, 1938) and proposed that meningeal vascular patterns can be used to sort them into different groups (Saban, 1983, 1985, 1986). However, these studies lack adequate comparative data regarding meningeal patterns in monkeys and apes. According to Diamond (19881, rhesus macaques (Macaca mulatta) retain a relatively conservative endocranial vascular configuration, and therefore provide a reasonable comparative model for the vascular state from which ape and human arterial patterns are derived. The purpose of the present paper is to present new information about meningeal arteries determined from a
large sample of rhesus monkey endocasts, and to discuss the implications of these data for understanding the evolution of cranial vasculature in apes and humans. MATERIALS AND METHODS Meningeal arterial patterns were studied from endocasts that were prepared by D.F. from skulls of rhesus monkeys (Macaca mulatta), using standard techniques (Falk, 1978).All observations were scored independently by both authors, and only specimens for which there was agreement were included in our sample of 200 hemispheres. In most cases, both hemispheres of an endocast were scored. Specimens represented both sexes and spanned from juveniles to adults. The relationship between meningeal arterial pattern and brain size was investigated, using cranial capacities collected by D.F. (accordingto standard methods; Falk, 1976) in conjunction with endocast preparation. Finally, referring to studies of direct dissections of pongid cranial vascular anatomy and human embryological data (Muller, 1977), we explored the implications of our findings for understanding the homologies between branches of the anterior meningeal arteries in human and nonhuman primates. The middle meningeal artery frequently enters the skull of rhesus monkeys through the lateral end of the foramen ovale (Castelli and Huelke, 1965a). However, Diamond (1988:57) observes two dissected specimens ofMacaca mulafta in which the artery “penetrates the cranial base by passing through the petrosphenoid fissure, posterior and lateral to the foramen ovale,” and reports that this condition exists in many cercopithecoids. Once the middle meningeal artery enters the skull, it divides into anterior (frontal) and posterior (parietal) branches. Although the subsequent course of the posterior branch is relatively constant in rhesus monkeys (Castelli and Huelke, 1965a1, the path taken by the anterior branch of the middle meningeal artery exhibits various patterns, which we scored from our endocasts. Diamond has adopted a system of nomenclature for the meningeal arteries that is based on stapedial artery homologies determined from embryological studies as well as comparison of a wide spectrum of primates
MENINGEAL ARTERLIES IN MACAQUES
and selected species from other orders (1988; 1991a,b). Diamonds system therefore differs from conventional nomenclature in that arteries are labeled according to their developmentaVphylogenetic histories rather than their origin, distribution, and function in postnatal individuals and extant species. Thus, if Diamond’s system were applied to macaques, the anterior meningeal artery that originates (via other branches) from the internal carotid artery would have the same name (“meningeal branch of the anterior division of the ramus superior of the stapedial artery”) as the anterior branch of the middle meningeal artery that originates from the maxillary artery of the external carotid artery, because he believes both of these arteries developed from the same branch of the stapedial artery (see, however, Muller, 1977). Although Diamond’s system would be appropriate for studies that focus strictly on homologies, it is inappropriate for the purpose of the present study, which is to identify and establish the range of variation in postnatal branching patterns of the meningeal arteries of one extant species of higher primate. Moreover, adoption of Diamond’s nomenclature would also result in a confusing and cumbersome terminology that is incompatible with that used in the wider literature on primate vascular anatomy. This is not to say that we discount the importance of embryological studies or that we think vascular homologies are unimportant. Indeed, we use these concepts to assess the implications of the findings for rhesus monkeys for vascular evolution in apes and humans. However, for reasons given above, we do not advocate changing the names of vascular structures to reflect their embryological or evolutionary histories. The scoring procedures used in this study were based on anatomical studies of cranial vasculature in macaques. These studies from the literature relied on direct dissection of cadavers, dissection of latex-injected specimens, corrosion preparations, and cleared specimens (Castelli and Huelke, 1965a,b; Weinstein and Hedges, 1962). Illustrations of three main patterns of distribution for the anterior branch of the middle meningeal artery, determined from 20 specimens by Castelli and Huelke (1965a), were
301
used to score meningeal artery patterns from rhesus endocasts. In the first pattern, the anterior division of the middle meningeal artery is reduced to a small anastomotic connection, which joins the anterior meningeal artery (which is off of the ophthalmic branch of the internal carotid artery) that supplies the side of the skull above the middle cranial fossa (see Fig. 2A). In the second pattern, the anterior meningeal artery and the anterior branch of the middle meningeal artery both provide branches to the side of the skull above the middle cranial fossa. These branches may be separate, or united t o varying degrees (Fig. 2B-D). Finally, in Castelli and Huelke’s third pattern, the middle meningeal artery supplies most of the sides of the anterior and middle cranial fossae while the anterior meningeal artery appears to be very small or absent (Fig. 2E). Below, we report the frequencies of these three patterns and two additional variations in a large sample of rhesus endocasts. We also explore the association between these patterns and cranial capacity (brain size). Because the origin of the middle branch of the middle meningeal artery has been a point of focus in the literature on human meningeal patterns, we scored that trait in the subset of rhesus endocasts that had the potential for having identifiable middle branches, namely those with patterns represented by Figure 2C,E. (The remaining patterns did not lend theniselves to such scoring because they lacked unambiguous extensions of a separate anterior branch of the middle meningeal artery onto the frontal and parietal surfaces.) Thus, using the system developed for humans by Adachi (1928), a hemisphere was scored as 0 if its middle meningeal artery lacked a middle branch; type I if a middle branch stemmed from the anterior branch of the middle meningeal vessels; type I1 if the middle branch was off of the posterior branch; and type I11 if clear middle branches stemmed from both anterior and posterior branches of the middle meningeal vessels (Fig. 1). RESULTS We observed five basic patterns for branching of the anterior division of the middle meningeal artery and its relationship to the anterior meningeal artery in the
302
D. FALK AND P. NICHOLLS
Fig. 1. Branching patterns of the middle meningeal artery that are frequently used to score humans, right sides. Type I, middle branch (Mmj arises from the anterior branch (Ma); Type 11, middle branch arises from the posterior branch (Mp); Type 111, middle branches arise from both anterior and posterior branches of the middle meningeal artery. Classification after Adachi (1928); illustration modified from von Bonin (1963).
200 rhesus monkey hemispheres (Fig. 2AE). The frequencies for each pattern are presented separately for right and left hemispheres in Table 1. The most frequent pattern (Fig. 2A) occurred in 69% of the endocast hemispheres. In this pattern, the anterior (or frontal) branch of the middle meningeal artery is truncated or reduced to a small connection at the junction with the anterior meningeal artery that supplies the side of the skull and dura above the middle cranial fossa (Castelli and Huelke, 1965a). Patterns R, C, and D in Figure 2 are all variations of the second most common pattern reported from dissection material (Castelli and Huelke, 1965a1, in which the anterior meningeal artery and the anterior branch of the middle meningeal artery both supply the side of the skull above the middle cranial fossa. That area is supplied by a common stem that is formed by the two arteries in pattern B, which is very rare in our sample. In pattern C, the two arteries are separated along their entire lengths, with the anterior branch of the middle meningeal running more or less parallel with and caudal to the anterior meningeal. (In one interesting variant of this pattern, the anterior meningeal artery crossed over the anterior branch of the middle meningeal.) Instead of running parallel, the two branches in pattern D join near their proxi-
mal ends, forming a triangle. Patterns B, C, and D together form 24.5% of our endocast sample. Finally, in pattern E (6.5% of the sample), the middle meningeal artery supplies most of the sides of the anterior and middle cranial fossae while the anterior meningeal artery appears to be absent. The posterior or parietal branch of the middle meningeal artery supplies the posterior part of the middle cranial fossa (frequently traveling with the petrosquamous sinus) and then sends a branch onto the posterior cranial vault and dura above the petrous portion of the temporal bone. We confirmed Castelli and Huelke’s (1965a) report that both the position and occurrence of this artery are relatively constant. However, we also observed that the extent of branching on the cranial vault varies from extremely simple t o elaborate, depending on the extent and elaboration of the posterior meningeal artery, which enters the skull near the junction of the transverse and sigrnoid sinuses. If the posterior meningeal artery is elaborate and extends rostrally across the parietal bone, the posterior branch of the middle meningeal artery tends to be modest, and vice versa (Fig. 3). Interestingly, Diamond (1988) observed a similar relationship for these two arteries in platyrrhine monkeys. We applied Adachi’s (1928) system for scoring branching patterns of the middle
MENINGEAL ARTERIES IN MACAQUES
D
303
E
Fig. 2. The five basic branching patterns of the anterior division of the middle meningeal artery and its relationship to the anterior meningeal artery in rhesus monkey endocasts, right sides. A, anterior meningeal artery; P, posterior meningeal artery; M, middle meningeal artery; Ma, anterior branch of the middle meningeal artery; Mp, posterior branch of the middle meningeal artery. The most frequent pattern observed was A, which appeared in 69%of 200 hemispheres. See text for other details.
meningeal artery in humans (described above) to the branching patterns of the 13 hemispheres that exhibited pattern E and to 38 of the 41 hemispheres with pattern C . For
the latter pattern, a middle branch originated off of the anterior branch of the middle meningeal artery (type I) in 8 specimens, off of the posterior branch (type 11) in 8 speci-
304
D. FALK AND P. NICHOLLS
anterior middle cranial fossa are supplied only by the anterior meningeal artery (patrhesus monkev endocasts tern A on both sides) than in specimens Right side Total Left side where both sides are supplied solely or parn n n (961 Pattern tially by the anterior branch of the middle 75 138 (69.0) A 63 B 01 02 03 (01.5) meningeal pattern (i.e., the group of en18 41 (20.5) docasts whose two sides exhibit any combiC 23 1) 04 01 05 (02.5) nation of B, C , D, and El. In the former 04 13 (06.5) E 09 group, cranial capacity averages 100.4 cm3 Total 100 100 200 (100) (range = 85-130 cm'; n = 471, whereas the mean for the latter group is 92.8 cm3 (range = 70-113 cm3; n = 12). This differmen&, f i ~ mboth branches (type III) in 8 ence in cranial capacity is statistically sigspecimens,andfromneitherbranchin 14hemi- nificant, t(57) = 2.20, P = .031, two-tailed. spheres. The ratings for pattern E were much Thus, there is a positive relationship besimpler. In all 13 hemispheres, the middle tween brain size and increased reliance on branch originated from the anterior branch the anterior meningeal and therefore internal carotid artery (i.e., when A appears on of the middle meningeal artery (type I). To determine if meningeal arterial pat- both sides) for supplying the dura and craterns are differentially represented in the nial vault above the middle cranial fossa. two sexes, we performed a chi-square test on Because the mean ages of the two groups endocasts drawn from our entire sample does not differ significantly (6.87 years for (with three categories: A, B C + D, and E). the first group consisting of pattern A on The result was x2 = 4.96, df = 2, which was both sides, 7.04 years for the group of ennot statistically significant. Thus, there are docasts with any combination of B, C , D, and no statistically significant sex differences in E, t < 11, the significant relationship bethe manifestation of meningeal arterial pat- tween meningeal pattern and cranial capacterns in the macaques we sampled. We also ity does not appear to be due to age. performed a similar chi-square test to determine if the different patterns were differenDISCUSSION AND CONCLUSIONS tially represented in the left and right hemiAlthough our observations of endocasts spheres across our entire sample. The result was x2 = 3.97, df = 2, which was not statis- from rhesus monkeys confirmed reports tically significant. A third chi-square test that meningeal arteries and veins frewas performed on two categories, A versus quently travel together, our investigation foB + C + D (some of the cells for E were cused on the arteries because of the availempty), to determine if there is a depen- ability of their descriptions from direct dence between patterns in the left and right dissections of Macaca mulatta (Castelli and hemispherees within those 73 individuals Huelke, 1965a). Indeed, we would not have for whom these meningeal patterns were been able to accurately interpret meningeal available from both hemispheres. The result arterial patterns from rhesus endocasts was x2 = 2.12, df = 1,which was not statis- without such reports. For example, had we tically significant. We therefore conclude relied on the literature for human that meningeal patterns are not differen- meningeal patterns, we would have misintially represented in left and right hemi- terpreted the most frequent rhesus monkey spheres and that, within individual speci- pattern (A) by erroneously assuming that mens, the pattern of one hemisphere does the anterior meningeal artery was the antenot predict the pattern of the other. (Inter- rior branch of the middle meningeal artery. estingly, Diamond (1988) reached another (This is because the location and visual configuration with respect to the anterior diviconclusion for hominoids.) We found that cranial capacity is signifi- sion of the middle meningeal artery superficantly larger in those endocasts in which the cially resembles the human condition.) The dura and both sides of the skull above the importance of interpreting meningeal patTABLE I , Frequencies of the fiue basic patterns
of the middle meningeal artery determined from
+
MENINGEAL ARTERIES IN MACAQUES
305
tends to be invcrse!g rclatcd t o Fig 3 The extent of branching ofthe posterior memnge?! srtery the amount of branching of the posterior branch of the middle meningeal artery (Mp), as illustrated by left and right posterior posterolateral views taken from two different rhesus monkey endocasts Lambdoid sutures and midlines are represented by dashes See text for discussion
terns on endocasts in light of direct dissections of appropriate species cannot be overstressed. Because of our large sample size, we were able to confirm and elaborate the trends and patterns observed by Castelli and Huelke (1965a) from their dissections of 40 right and left sides combined. Castelli and Huelke’s most frequent pattern occurred in 52% of the sides and is equivalent to the most frequent pattern (A) that occurred in our sample (69% of 200 hemispheres). Our study extends the findings of Castelli and Huelke by reporting frequencies for the individual variations of their second most frequent pattern (28% of their sample). Thus, patterns B, C , and D are all variations of this pattern and together total 24.5% of our sample. Finally, our pattern E (6.5%) was the least frequent of the three patterns that Castelli and Huelke described (20%). As the above shows, the overall distribution of the frequencies of the three meningeal patterns that we observed in our endocasts (n = 138, 49, and 13 for Castelli and Huelke’s Patterns 1, 2 and 3, respectively, yielding proportions of .69, .245, and .065) is similar to the overall distribution that Castelli and Huelke observed in direct dissections of macaque cerebral arteries (n = 21, 11, and 8 for Patterns 1, 2 and 3, respectively, yielding proportions of 5 2 , .28, and .20). Nevertheless, the patterns of our and Castelli and Huelke’s overall frequency distributions were statistically different
(x2= 8.43, df = 2, P < .02). It appears that this overall significant difference was largely due to our observing a greater proportion of Pattern 1(or A) but a smaller proportion of Pattern 3 (or E) than Castelli and Huelke observed. It is possible that some endocasts that actually represented pattern E were misidentified as belonging to pattern A. This could occur if the anterior meningeal vessels that normally appear as pattern E on an endocast had been located somewhat rostrally and were buried within the anterior end of the Sylvian fissure in the living monkey. Despite this possible “noise”in our data, we are satisfied that our overall results reveal the trends predicted from the dissection data, and provide additional details about meningeal arterial patterns in rhesus monkeys. We also note that, although their frequency may be underestimated for the above reasons, the set of endocasts classified as manifesting pattern E does not appear to contain specimens that have been misidentified. As shown in the results section, our exercise in applying Adachi’s method for classifying the origin of the middle meningeal artery to rhesus monkey endocasts did not work very well for pattern C, since 14 of 38 hemispheres did not even appear to have a middle branch and the other 24 were evenly distributed across the other three possible conditions. On the other hand, all 13 of the hemispheres manifesting pattern E could be scored. A middle branch originated off of the
306
D. FALK AND P. NICHOLLS
anterior branch of the middle meningeal artery in all of them. We can therefore conclude that in rhesus monkeys where there is a predominance of the middle meningeal artery over the anterior meningeal artery (i.e., on the side of the skull above the middle cranial fossa), this pattern is associated with branching of the anterior division of the middle meningeal artery. It would be premature to compare these data for rhesus monkeys with those for humans, however, without first establishing the ~7ariouspatterns and their frequencies for meningeal arteries in the great apes. The fact that the group of endocasts that relied solely on the anterior meningeal artery to supply the side of the skull and dura mater above the anterior part of the middle cranial fossa (pattern A on both sides) have a significantly larger mean cranial capacity than the group o f remaining endocasts shows that increased brain size in rhesus monkeys is associated with increased dependence on the internal carotid artery (from whence stems the ophthalmic artery that gives rise to the anterior meningeal artery). It would be interesting to determine whether our findings based on intraspecific variation in rhesus monkeys bear on a phylogenetic analysis. In other words, do the great apes (who are bigger brained than rhesus monkeys) rely on branches of the ophthalmic artery rather than the middle meningeal artery lo supply the sides of the skull above the middle cranial fossae? Direct dissections o f cranial arteries in one gorilla and one orangutan in conjunction with observations of grooves of meningeal arteries in skulls of two additional gorillas suggest that the answer to this question is yes (Muller, 1977). In these pongids, the most rostral portion of the sides of the skull above the middle cranial fossae are supplied by branches from the meningeal lacrimal artery, which is a branch of the ophthalmic artery that stems from the internal carotid artery. (Muller observes that these extra-orbital branches of the lacrimal artery enter the braincase through the supra-orbital fissure or, alternatively, through the cranio-orbital foramen.) The middle meningeal vessels are distinct from and caudal to these branches. Although Dia-
mond (1991a) disagrees with Muller’s conclusions regarding homologies, his own dissection of one chimpanzee specimen revealed that, on both sides, the meningeal branch to the frontoparietal area was “fed in retrograde fashion by the ophthalmicthalmic artery” (p. 235). Furthermore, Diamond notes that on the right side of one of two orangutan specimens that he dissected, “the definitive lacrimal artery gave off a recurrent branch that entered the middle cranial fossn through 3 foramer, . . .” fp. 23;;. (Although discussion of the controversies surrounding meningeal-orbital arteries is beyond the scope of this paper, this topic is reviewed by Diamond, 1991a.) The pattern reported by Muller (1977) is similar to the most common pattern reported from rhesus monkey dissections, in which the anterior meningeal artery occupies the same rostral location and also stems from the ophthalmic artery. (Compare Fig. 1 of Castelli and Huelke, 1965a with Fig. 5 of Muller, 1977; see Fig. 4.) Thus the meningeal lacrimal artery of pongids and the anterior meningeal artery of rhesus monkeys appear to be likely homologues. Muller (1977) proposes that the anterior branch of the middle meningeal artery in humans is the homologue of the pongid meningeal lacrimal artery, based on a study of cranial vascular development in a series of injected human fetuses that varied from 143 to 290 mm crown-rump length. During fetal development, the lacrimal artery (which is a branch of the ophthalmic) enters the middle cranial fossa from the orbit where it connects with the middle meningeal artery medially, while delivering a named branch (rml, the principal meningeal branch of the lacrimal artery) onto the wall of the cranium laterally. Rml ascends the vault between the frontal and parietal bones, and (as development proceeds) the middle meningeal artery eventually takes it over (“annexes”it) through the connection mentioned above. Thus, the anterior branch of the middle meningeal artery of humans appears homologous not only with the meningeal lacrimal artery of pongids (Muller, 1977), but also with the anterior meningeal artery of rhesus monkeys. (For an alternative view, see Dia-
MENINGEAL ARTERJES IN MACAQUES
307
ference is that the homologous artery is usually distinct from the middle meningeal artery in monkeys and apes, but has become incorporated as the anterior branch of the middle meningeal artery in humans. Muller’s comparative study of apes and humans therefore led to the conclusion that the “middle meningeal artery gains in importance as mammalian evolution proceeds” (1977:218).However, it remains to be seen if and how the patterns described above for rhesus monkeys are distributed within and across samples of great apes, or if apes manifest additional arterial variations. We hope to pursue these issues in future studies and to apply our findings to furthering the understanding of the evolution of cranial vasculature in monkeys, apes, and humans. ACKNOWLEDGMENTS We thank Marc Schmidt for help with illustrations and are grateful to Jim Neely and Michael Diamond for comments on the manuscript. The University of Puerto Rico is acknowledged for providing free access to the Cayo Santiago skeletal collection. This research was supported by NSF grant No. BNS-9008179.
I Fig. 4. Most frequent arterial branching pattern observed in rhesus monkey endocasts (above) compared to that of a 3 year old gorilla (middle) and an adult gorilla (below), right sides. The gorilla patterns were determined by Muller (1977). In apes, the most rostra1 portion of the sides of the skull above the middle cranial fossae are supplied by branches of the meningeal lacrimal artery (La), which is a branch of the ophthalmic artery. La appears to be homologous with the anterior meningeal artery (A) of rhesus monkeys because of its location, origin, and configuration with respect to the anterior branch of the middle meningeal artery. See text for discussion and Figure 2 for other abbreviations.
mond, 1991a.) All three of the proposed homologues derive (developmentally) from the ophthalmic artery, and supply the same regions of the skull and dura mater. The dif-
LITERATURE CITED
Adachi B (1928)Das Arteriensystem der Japaner. Band 1. Kyoto: Verlag der Kaiserlich-Japanischen Universitat zu Kyoto. Batson OV !1944) Anatomical problems concerned in the study of cerebral blood flow. Fed. Proc. 3r139-144. Bonin G von (1963) The Evolution of the Human Brain. Chicago: The University of Chicago Press. Castelli WA, and Huelke DF (1965a)The arterial supply of the dura mater of the rhesus monkey. Anat. Rec. 152155-160. Castelli WA, and Huelke DF (1965b) The arterial system of the head and neck of the rhesus monkey with emphasis on the external carotid system. Am. J. Anat. I 1 6;149-170. Crosby EC, Humphrey T. and Lauer EW (1962) Correlative Anatomy of the Nervous System. New York: Macmillan. Dahl E, and Edvinsson L (1988) Anatomical organization of cerebral and extracerebral vasculature. In J Olesen and L Edvinsson (eds.): Basic Mechanisms of Headache. Amsterdam, Elsevier Sci. Pub., pp. 2 7 4 7 . Diamond MK (1988) Cephalic Vascular Evolution and Development in Primates: The Stapedial Artery and its Companion Venous Sinuses. PhD dissertation, University of Chicago. Ann Arbor: University Microfilms.
308
D. FALK AND P. NICHOLLS
Diamond MK (1991a1 Homologies of the meningeal-orbital arteries of humans: A reappraisal. J . Anat. 17#:223-241. Diamond MK (1991b)Homologies of the stapedial artery in humans, with a reconstruction of the primitive stapedial artery configuration of euprimates. Am. J. Phys. Anthropol. #4:433-462. Falk D (1976)External Neuroanatomy of the Cercopithecoidea. PhD dissertation, University of Michigan. Ann Arbor: University Microfilms. Falk D (19781 External Neuroanatomy of Old World Monkeys (Cei-copithecoidea).Contrib. Primatol. 15t195. Jones FW (1912) On the grooves upon the ossa parietalin rommonly said to he raiised hy the arteria moningea media. J. Anat. 46:22&238. Muller F (1977) The development of the anterior falcate and lacrimal arteries in the human. Anat. Embryol. 150;207-227.
Padget DH (1956) The cranial venous system in man in reference to development, adult configuration, and relation to the arteries. Am. J. Anat. 98t307-355. Saban R (1983) Les veines meningees moyennes des australopitheques. Bull. Mem. SOC.d’Anthrop. (Paris) 1Ot313-324. Saban R (1985) Identification des stades evolutifs des hommes fossiles d’apres les veines meningees. Actual. Odonto-Stomatol. 152t693-712. Saban R (1986) Veines meningees et hominisation. Anthropos (Brno)23r15-33. Weidenreich F (1938) The ramification of the middle meningeal artery in fossil hominids and its bearing upon phylogenetic problems. Palaeontologia Sinica New Sw D.?:l-l6 Weinstein JD, and Hedges TR (1962) Studies of intracranial and orbital vasculature of the rhesus monkey (Macaca mulatta).Anat. Rec. 144r37-41.
