AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 81167-76(1990)

Position and Orientation of the Foramen Magnum in Higher Primates S.A. LUBOGA AND B.A. WOOD Department of Anatomy, Makerere University, Kampala, Uganda (S.A.L.); Hominid Palaeontology Research Group, Department of Human Anatomy and Cell Biology, The University of Liverpool, England

KEY WORDS

Skull base, Evolution, Variation, Allometry

ABSTRACT

The location of the foramen magnum, with respect to the longitudinal axis of the cranium, and its orientation with respect to the Frankfurt Horizontal, have been studied in a total of 328 modern human and Pun crania. The samples were chosen in order t o examine the effect of overall size difference on foramen magnum disposition. Foramen position (expressed as three indices) and inclination are relatively invariant among the modern human samples, but the foramen magnum is consistently, and statistically significantly, more anteriorly located in Pun puniscus than in Pun troglodytes. Sexual dimorphism is virtually non-existent. There is an apparent allometric effect on foramen position, but not on inclination, so that larger crania in the modern human and Pun puniscus samples tend to have more posteriorly situated foramina. The disposition of the foramen is unrelated to cranial base angle or facial prognathism, except that in Pun puniscus its relative anterior location is linked with the more flexed cranial base in that species. These results provide a comparative context for the examination of differences in foramen magnum disposition in fossil hominids. Differences in foramen magnum position and orientation between KNM-ER 1813 and A. ufricanus are most unlikely to be due to within-taxon variability.

The position and orientation of the foramen magnum have come to assume a special significance for those who seek to interpret the taxonomic and functional significance of fossil hominid cranial remains. Whereas other studies (Daubenton, 1764; Broca, 1875; Welcker, 1862; Huxley, 1863) dwelt on the significance of either the position or shape of the foramen magnum, Topinard (1878) was the first to put its anterior position and horizontal orientation in modern humans into a sound, but essentially descriptive, comparative context. It was Bolk, however, who was the pioneer in providing quantitative data about both the position (Bolk, 1909) and orientation (Bolk, 1910) of the foramen magnum. He related basion (the anterior border of the foramen magnum in the sagittal plane) to a line joining the anterior (fronton) and posterior (occipiton) extremities of the endocranial cavity as the components of a “basal index”; the smaller the index, the more anteriorly situated the

@ 1990 WILEY-LISS. INC

foramen magnum. Bolk used the same “fronton to occipiton” line as the reference plane for comparisons of foramen magnum orientation. When he studied a wide range of non-human primates, he found an approximate correlation between basal index values and foramen angle. At one end of the spectrum were modern humans, in which the foramen was both anteriorly situated and horizontal, whereas inMycetes seniculus (Alouuttu seniculus-the Red Howler monkey), the foramen is both posteriorly situated and close to vertical. However, within the nonhuman apes he studied, no such correlation was evident, with Pongo having the most anteriorly situated, yet the least horizontal, foramen magnum. Bolk (1915)also noted the similarity between the disposition of the foReceived August 30,1988; accepted February 22,1989. Address reprint requests to Department of Human Anatomy and Cell Biology, The University of Liverpool, P.O. Box 147, Liverpool, L69 3BX, U.K.

68

S.A. LUBOGA AND B.A. WOOD

ramen in juvenile ape and adult modern human crania. Senyurek (1938) and Schultz (1942,1955) used the location of the occipital condyles as a “marker” for the position of the foramen magnum, the latter author combining them in a “condylion index” (nasion-condyliod nasion-opisthocranion . 100). Schultz’s results confirmed those of Bolk, which were based on the “basal index.” Schultz (1955) also showed that, of the non-human apes, Pan paniscus came closest to the values recorded for modern human crania. Le Gros Clark (1950)related the position of the occipital condyles to a different horizontal reference line (the maximum anteroposterior projection in the Frankfurt Horizontal) in yet a third index, the “condylar position index.” The smaller the index value, the more posteriorly located the condyles. In Gorilla and Pan troglodytes, the mean values are comparable (24 and 23, respectively), but the mean value for the five crania of Pan paniscus suggests that the condyles, and thus the foramen, are more anteriorly situated in the pygmy chimpanzee. Similar results were obtained by Gmbel et al. (1984) on a small sample of Pan paniscus, using sellion and prosthion as the anterior termini and both opisthion and basion as indicators of foramen magnum position. Cramer (1977) has also commented on the peculiarities of Pan paniscus foramen magnum position. Adams and Moore (1975) employed a similar combination of variables and concluded that the weight of ape jaws was not balanced by bony factors alone, and they inferred that the nuchal musculature is more effective, per unit area of insertion, in the apes than in Homo sapiens. Others (Ashton and Spence, 1958; Ashton and Zuckerman 1951, 1952, 1956) have studied the position of the foramen in adults as well as using cross-sectional ontogenetic data. Delattre and Fenart (1963) and Matsumoto (1983, 1986) have related cranial landmarks in multidimensional space and included variables that would be affected by the position of the foramen magnum, but make no special point of discussing this variation. Dart (1925) was sufficiently confident about the significance of foramen magnum morphology to speculate about the posture of the Taung infant on the basis of no more than the disposition of its foramen magnum. He did this by relating basion to the prosthioninion line in a “head-balancing” index. Wei-

denreich (19431,Broom and Schepers (1946), Le Gros Clark (1967), Tobias (19671, and Kimbel et al., (1984) have all included evidence from the foramen magnum in their analyses of fossil hominid remains. Tobias (1967) provides a summary of his own and previous work that uses the position and the inclination of the foramen magnum to contrast hominids and pongids as well as to make comparisons between hominids. However, the comparative pongid samples are meagre: Tobias’ amounted to a single female gorilla skull and Weidenreich (1943) illustrates just one cranium each of Gorilla, Pan, andPongo (figs. 188-190) as his comparative data. Even so, it is noteworthy that Weidenreich (1943) commented on the lack of variation in the pongid data. However, it is unusual for the foramen magnum to be preserved in fossil hominid crania, and the occipital condyles have been used as a guide to the location of the foramen magnum, but these too are vulnerable to damage. Others have used the position of the external auditory meatus (Wood, 1976a) to estimate the location of the foramen magnum with respect to the sagittal axis of the cranium. Such measures are inevitably crude, but before even their value can be assessed, more comparative information about variation in both foramen magnum position and orientation is required. How variable, for example, are the latter both within and between modern human and nonhuman ape taxa? Does the position of the foramen in relation to the long axis of the cranium have a predictable relationship with overall cranial size? Finally, is there any evidence, either within or between taxa, of a correlation between facial prognathism, cranial base flexion, and the disposition of the foramen magnum? This study was not so much concerned with variability of foramen magnum position and orientation among taxa, but was designed to explore variation within closely related taxa. Attention was paid to variability within Pan, because the two extant species, Pan troglodytes and Pan paniscus, provide an excellent opportunity to study any allometric effect using the concept of “narrow” allometry (Smith, 1980). This is because the two species of Pan are nearly identical genetically (Hasegawa et al., 1987; Sibley and Ahlquist, 19871,yet differ enough in body size to be suitable for allometric study. The presence o r absence of allometric

69

HIGHER PRIMATE FORAMEN MAGNUM

influences were investigated in the modern human sample, and sexual differences in foramen magnum location and orientation were also investigated.

could be measured on only 204 of the modern human crania. Repeated measurements of both the four linear and the single angular measurement were made to test for measurement error; differences between the sets MATERIALS AND METHODS of measurements were consistently less than Linear measurements and angles were re- 5%of the smallest value. corded from a total of 328 modern human The allometric relationship between the and Pan crania; the sample is detailed in position of the foramen magnum and overall Table 1. All specimens were judged to be cranial size was investigated by comparing adult on the basis of both the alveolar erup- the major axis slopes of the log-transformed tion of the third molar and the fusion of the distances BS-FC (5),BS-NA (61, and BS-SN spheno-occipital synchondrosis. The sex of (71, each regressed on the logarithm of the each cranium was either taken from the geometric mean (GM) of the linear measurerecords of the collection or deduced on the ments of each cranium (Luboga, 1986). basis of osteological characters (e.g., Lar- These three relationships were investigated nach and Freedman, 1964; De Villiers, 1968; in two data sets, both of which qualify as Brothwell, 1981). Diseased or badly dam- “narrow” allometry studies, as defined by aged crania were not included, nor were Smith (1980). The data sets were (A) the specimens that showed evidence of antemor- combined sample of modern human crania tem tooth loss. (N = 240) and (B)the combined sample of the The following landmarks were located on two Pan species (N = 88). In each case, the both the original crania and on specially three “test” coefficients are compared with prepared lateral radiographs of each speci- the major axis regression coefficient bemen: glabella (GL), nasion (NA), foramen tween log overall cranial size and the logacaecum (FC), sella (S),subnasale (SN), alve- rithm of cranial length (OPN-GL).Allometolare (AL), opisthocranion (OPN), opisthion ric relationships within and between taxa (OP), and basion (BS). All landmarks were were assessed by computing and comparing projected onto the Frankfurt Horizontal the major axis slopes. Formulae for their (FH) as shown in Figure 1.Definitions of the calculation, and for the calculation of confireference points and details of radiographic dence limits, are given in Luboga (1986). technique are given in Luboga (1986). The Overall size of the cranium was taken to be following linear measurements were taken: the geometric mean of the 41 linear mea1)OPN-GL, 2) OPN-SN, 3) OPN-FC, and 4) surements of the cranium used in a larger OPN-BS (Fig. 1).The position of the foramen study (Luboga, 1986). All data were conmagnum, as represented by point (BS), and verted to natural logarithms. Cranial base flexion in the sagittal plane of the other landmarks was located with respect to the FH and expressed in the form was measured by the angle BS-S-FC, and of three indices: 1)OPN-BS/OPN-FC 100; maxillary alveolar and basal prognathism 2) OPN-BS/OPN-SN .loo; and 3) OPN-BS/ (Bjork, 1950) were estimated using the anOPN-GL. 100. The angle of the foramen gles S-FC-ALand S-FC-SN (Fig. 2). magnum (FM) was measured on the lateral RESULTS radiographs and was taken to be the angle The means, standard deviations, coeffibetween the projection of a line joining opisthion (OP) and basion (BS) and the cients of variation (where applicable), and Frankfurt Horizontal (Fig. 2); the angle ranges of the three foramen magnum posiTABLE 1. Crania used in the study, by taxa and sex Taxon -

~

-

Homo sapiens African Chinese Romano-British Australian aboriginal Pan troglodytes Pan paniscus

Male ~

~

41

43 32 45 21

20

Female ~

~~

27 8 28 16

25 22

Total ~

68 51 60 61 46

42

~

~~~

~

~

Source-~ ~

- -

Galloway Collection, Makerere B.M. (N.H.) B.M. (N.H.) Poundbury B.M. (N.H.) and Department of Anatomy, Edinburgh Powell-Cotton Collection, Birchington Musee Royal de 1’Afrique Centrale, Tervuren

70

S.A. LUBOGA AND B.A. WOOD

,

I

I

r--I I

--

1

Fig. 1. Landmarks used in the study, together with the four linear measurements derived from them when they are projected onto the Frankfurt Horizontal.

Fig. 2. Illustration of the angle of inclination of the foramen magnum (FM) (i,e., the angle between the plane OP-BS and the Frankfurt Horizontal), and the angles used to measure cranial base flexion (BS-F-FC) and facial prognathism (S-FC-SN and S-FC-ALV).

tion indices for four geographically distinct samples of modern human crania are given in Table 2. Pairwise comparisons were made among the four groups for each of the three indices using a two-tailed Student’s t-test. None of the differences between the geo-

graphical subsamples were significant ( P < .05). Pairwise testing also confirmed that sexual differences were few and not statistically significant. Foramen magnum orientation for the modern human samples is also presented in

71

HIGHER PRIMATE FORAMEN MAGNUM

TABLE 2. Position and orientation of the foramen magnum in separate-sex samples Samples

SD

Mean

~~

~~

African

~~~

(1)' (2) (3)

Angle FM3 Chinese Angle FM RomanoBritish Angle FM Australian aborigine

Male CV

~

53.4 48.8 51.8 4.9 54.9 49.5 51.5 7.3 55.3 50.9 52.8 9.8 51.4 47.4 51.2 7.6

(1) (2) (3) (1) (2) (3) (1) (2) (3)

Angle FM

~

2.6 2.0 2.0 4.3 3.9 2.5 2.3 4.5 3.4 2.7 2.2 4.2 3.0 2.0 2.1 5.3

Female ~-Range ~

4.9 4.0 3.9

7.1 5.0 4.5 -

6.1 5.3 4.1

5.8 4.2 4.1

-

modern human crania

of

47.9-60.1 44.8-54.7 47.2-57.4 -6.2-13.6 47.0-63.5 45.1-57.3 47.2-56.6 -2.4-15.8 48.4-61.6 45.9-55.6 48.0-56.5 2.4-19.2 40.5-57.0 42.2-51.8 43.6-55.0 -5.1-17.3

Mean ~~

CV

SD ~~

52.9 48.5 51.6 8.4 55.5 49.8 51.5 7.6 55.3 51.9 53.5 9.4 51.7 46.9 51.0 8.8

Range ~

2.8 1.5 1.7 5.0 4.3 2.1 3.1 6.6 2.5 2.0 1.7 5.0 2.1 2.1 1.9 4.0

5.3 3.0 3.2

7.7 4.2 6.0

4.5 3.9 3.2

4.0 4.5 3.7

-

~

~~

46.0-57.8 46.2-51.4 49.6-55.0 -2.9-17.7 48.8-61.5 45.2-51.6 45.9-55.1 0-17.6 50.3-60.2 48.3-57.1 50.0-58.9 2.5-20.2 47.5-55.4 43.1-50.8 48.2-54.4 0-14.0

%SD2 99 99 100

101 101 100

100 102 101

101 99 100

-

'(1) = OPN-BS/OPN-FC . 100; (2) = OPN-BS/OPN-SN . 100; (3) = OPN-BS/OPN-GL . 100. '% SD = female m e a n h a l e mean 100. "Foramen magnum angulation could be measured only on 204 of the total of 240 modern human crania available.

TABLE 3. Position and orientation of the foramen magnum in separate-sex samples of Pan Samples

Pan troglodytes Angle FM Pan paniscus Angle FM

___

(1)' (2) (3) (1) (2) (3)

Mean ~~~~

38.0* 29.5 38.l* -19.6 41.6 34.2 42.7 -19.1

SD 2.8 2.2 3.8 4.5 2.9 2.5 3.4 5.7

Male CV 7.4 7.5 10.0

7.0 7.3 8.0

-

Range 32.7-42.3 25.4-33.1 31.7-44.9 -26.7 to-8.7 37.5-48.9 31.3-40.0 37.2-49.6 -25.6 to -7.0

Female Mean___SD CV - ~. 38.0* 30.0* 38.0* -23.1 42.1 35.0 42.7 -15.0

3.1 2.5 3.5 4.9 2.4 2.4 3.2 5.8

8.2 8.3 9.2

5.7 6.9 7.5

-

Range -

%SD2

31.7-45.0 25.3-35.2 31.4-44.9 -33.5 to -14.6 38.9-46.9 31.7-40.3 36.2-49.6 -25.9 to -4.8

100 102 100

~~

101 102 100

-

OPN-BSIOPN-FC . 100; (2) = OPN-BS/OPN-SN . 100; (3) = OPN-BS/OPN-GI.. 100. %I SD = female meadmale mean 100. *Difference between the means of Pan troglodytes and Pan paniscus at P =