Endocranial suture closure in rhesus macaques (Macaca mulatta)

AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 80:417-428 (1989)

Endocranial Suture Closure in Rhesus Macaques (Macaca mulaita) DEAN FALK, LYLE KONIGSBERG, R. CRISS HELMKAMP, JAMES CHEVERUD, MICHAEL VANNIER, AND CHARLES HILDEBOLT Department of Anthropology, State University of New York,Albany, New York 12222 (D.F.), Departments of Anthropology (L.K.,J.C.) and Cell Biology and Anatomy (J.C.), Northwestern University, Evanston, Illinois 60208; Department of Sociology and Anthropology, Purdue University, West Lafayette, Indiana 47907 (R.C.H.); Mallinckrodt Institute of Radiology, Washington University School of Medicine, St. Louis,Missouri 63110 (M.V., C.H.)

KEY WORDS Age determination, Asymmetries, Cranium, Endocasts, Petalias ABSTRACT Endocasts from skulls of 330 rhesus monkeys (Mucaca muZatta)of known age are scored for closure of nine bilateral and three unilateral sutures or segments of sutures. A variety of tests reveals a strong relationship between age and stages of suture closure, although increasingly broad confidence intervals prevent sutures from being very useful for precisely aging older macaques. The order in which endosutures begin to close, as well as that in which closure is finally achieved, is determined for macaques, and these sequences compared to those for endosutures of humans (Todd and Lyon, 1924). The basilar suture is the earliest to close, while the masto-occipital and rostral and caudal squamosal sutures achieve closure quite late in both species. On the other hand, humans and macaques differ in their schedules for the sphenofrontal suture and in the initiation of closure for the rostral portion of the squamosal suture. Two sutures close significantly sooner on the right than on the left side (the rostral squamosal and masto-occipital) and asymmetry favoring closure of the right lateral lambdoid suture also approaches significance at the 0.05 level. No sutures close significantly sooner on the left side. It is suggested that macaque sutures may close from the inside out, that endosutures are more sensitive than ectosutures for detecting sequences in which cranial sutures begin to close, and that directional asymmetries in suture closure of macaques may be related to minor asymmetries in braid skull shape (petalias). The use of cranial suture closure to age human skeletal material has had a checkered history. Numerous workers have raised serious questions about using suture closure as an aging method (Singer, 1953; Brooks, 1955; McKern and Stewart, 1957). However, Meindl and Lovejoy (1985) have recently shown that a combination of segments of ectocranial sutures can contribute valuable information to age estimates. In this report, we explore the relationship between closure of endocranial sutures and age in rhesus monkeys (Macaca mulatta) and compare our results with those for both endocranial (Todd and Lyon, 1924) and ectocranial (Meindl and @ 1989 ALAN R. LISS, INC

Lovejoy, 1985) sutures of humans. The order of suture closure is compared in the two species. Because Zivanovic (1983) has noted a considerable amount of hemispheric asymmetry in suture closure among humans, we tested for directional asymmetry in suture closure in rhesus monkeys. Our findings may shed light on the development of cranial petalias in human (LeMay, 1977; Galaburda et al., 1978) and nonhuman (LeMay et al., 1982) primates. Received October 19, 1988; revision accepted November 14, 1988.

418

D. FALK ET AL.

MATERIALS AND METHODS

Latex endocranial casts (endocasts) were prepared by D.F. from 330 skulls of rhesus macaques from Cay0 Santiago. In some analyses, sample sizes are reduced from 330 because of missing data. Samples from this same collection have been used to develop aging criteria based on pubic symphysis development (Rawlins, 1975) and dental eruption and epiphyseal fusion (Cheverud, 1981). Animals’ages at death were known to within at most one year, with most ages at death known to within one week (Sade et al., 1985; Busse, personal communication).Endocasts, whose preparation is described in detail elsewhere (Radinsky, 1968; Falk, 1978), reproduce features imprinted on the endocranial surface, including clear representations of sutures. Each endocast was dipped in polyvinylacetate for reinforcement, and then sprayed with metallic gold paint to enhance visibility of surface features. Closure data were collected by RCH for nine bilateral sutures or segments of sutures including the masto-occipital, sphenofrontal, sphenotemporal, medial coronal, lateral coronal, medial lambdoid, lateral lambdoid, rostral squamosal, and caudal squamosal (called the parieto-occipital by some authors). Three unilateral sutures (or segments) were also scored: the basilar, rostral sagittal, and caudal sagittal (Fig. 1). Most of these features were reproduced on the 330 endocasts, although a few sutures were unobservable on some endocasts.

Fig. 1. Left lateral view of endocast from rhesus monkey skull illustrating the sutures scored in this study. Numbers correspond to sutures listed in Table 3: 1, medial coronal; 2, lateral coronal; 3, rostral sagittal; 4, caudal sagittal; 5, medial lambdoid; 6, lateral lambdoid; 7, masto-occipital;8, basilar; 9, sphenofrontal;10, sphenotemporal; 11, rostral squamosal; 12 caudal squamosal.

Suture closure was rated according to a modification of Fredric’s traditional scale after Krogman (1978). Sutures were scored as open (0)if there was sufficient separation of the adjacent cranial bones to permit liquid latex to flow through to the ectocranial surface along the entire length of the suture; closing (1) if the interdigitation of adjacent bones prevented passage of the liquid latex along a portion of the suture; closed (2) if interdigitation was complete along the entire suture thus blocking the flow of any latex to the ectocranial surface, and a t least a portion of the course of the endocranial suture margin was still evident from its impressions on the inner table; and obliterated (3)if all endocranial traces of the suture had been eliminated by fusion of the endocranial margins. Intraobserver error was measured by repeated scoring of all variables on 30 endocasts. Repeatability for individual sutures (segments) ranged from 90 to loo%,the average being 93.5%. These repeatability scores are in keeping with those determined for single variables in other suture studies (e.g., Meindl and Lovejoy, 1985). We examined the level of asymmetry in macaque suture closure to determine whether bilateral sutures can be combined or must be treated separately in an analysis of age regression. Although Zivanovic (1983) did not distinguish between fluctuating and directional asymmetry in suture closure, we make this distinction here, as it is central to the question of combining bilateral sutures. Directional asymmetry occures when a trait is typically more developed on one particular side than the other. Fluctuating asymmetry occurs when a trait differs from side to side, with neither side typically showing more development. Even under high levels of fluctuating asymmetry it may be possible to substitute an antimere for its partner, but with directional asymmetry an antimere will give biased age estimates using a standard developed on the opposite side. To examine the lefvright correlations for closure of bilateral sutures we used a common measure from an ordinal contingency table. The lefuright correlations express the relationship between sides, and are decreased by any fluctuating asymmetry. The ordinal contingency tables are formed by cross tabulating observations on the 330 casts for a left and right suture across the four ordered classes of closure (0, 1, 2, 3). Goodman and Kruskal’s (1954) gamma ( y )

419

ENDOSUTURES OF MACAQUES

can then be used t o measure the association between the left and right side. Gamma is defined as:

right and left sides with gamma, but uses delta to test for significant differences between sides, i.e. sides may be correlated but still be significantly different.) Delta is defined as: A = C C Pi+P+j- C C Pi+P+j

where C is the number of concordances and D the number of discordances across all pairs of individuals for the left and right suture. As an example of a concordance, if individual i has a score of 1 on the right suture and 2 on the left, then this individual’s ranking on the two sutures will be concordant if compared to an individual j with 0 on the right and 1 on the left. In other words, the rankings are concordant because individual i outranks individual j for each side. In a discordance, the sides rank inversely; for example, an individual scored as 2,l will be discordant with an individual scored as 1,2. Gamma varies from 1 (perfect concordance) to -1 (perfect discordance). “C” and “D’ can be easily calculated from the contingency table as: C= C

C nij nkl

i15 4-10 5-9 5-10 7 > 15 9 > 15 >15>15

No. of years 14 >I3 6 4 5 >8 >6

-

Duration (years)

30-66 3824-41 22-65 23-35 26-47 26-81 37 2 81

No. of years

HomdMacaca (No. of years)

36 17 43 12 21 55

2.57 2.83 10.75 2.40 12.63 59.17

544

-

'Age spans for macaques were determined from Table 3. For each suture, the beginning of closure is the age when 0.0% of the endocasta are still open (i.e., 0.0% score 0). The end of closure is the age when 0.0% of the endocasta are open and 0.0% are in the process of closing (i.e., 0.0%score 0 and 1). Ranges for sutures that were scored in segments (e.g., coronal) were determined by examining the data for both portions (medial and lateral). Durations for Homo are taken from Todd and Lyon (1924).

rial, which cannot be aged by dental eruption or epiphyseal union. Although our data demonstrate a strong relationship between stages of suture closure and age of macaques, they also show that sutures beDISCUSSION come progressively inefficient as age indicaOne of our objectives was to develop a tors with advancing age. While the ineffimethod for aging older rhesus skeletal mate- ciency of age estimates in the later age

macaques that are beginning to close in only one respect, the (lateral) coronal suture precedes the sphenotemporalto rostra1 squamosal sequence.

426

D. FALK ET AL. TABLE 6. Order of suture closure in humans and macaques’ Seauence Macaca endosutures Initiation Closure Homo endosutures Initiation Closure Homo ectosutures Initiation Closure

sptem-r/sq-cor-(spfro sag)-lb-masoc-c/sq spfro-(cor sag)-sptem-(lb masoc r/sq c/sq) spfro-sag-cor-(lb masoc)-sptem-c/sq-r/sq sag cor- lb spfro -sptem -masoc- c/sq -r/sq

-

-

sag-cor-lb-spfro-sptem sag-cor-lb-spfro-sptem

’Sequences in which cranial sutures begin to close (initiation) and those in which they achieve closure. Endosuture sequences for macaques and humans (Todd and Lyon, 1924) determined from Table 5. The sequences for human ectosutures determined from Meindl and Lovejoy’s (1985) Table 3 (counting their point 7, pterion, a s the lateral coronal) remained the same across all four of their suture scores. Although the methods vaned by which authors determined and quantified ages for suture closure, the overall sequences seem fairly comparable. Abbreviations: c/sq, caudal squamosal; cor, coronal; lb, lambdoid; masoc, masto-occipital; r/sq, rostra1 squamosal; sag, sagittal; spfro, sphenofrontal; sptem, sphenotemporal. Sutures contained in parentheses closed at the same time.

TABLE 7. Matrix of relationshim between suture closure states’ 8

6 7 12

50 60 43 55 29 26

2

10

1

11

9

4

138 0 33 36 41 24 65 59 45 24 9

140 44

134 54 55 0 52 41 55 47 48 29 10 3

136 61 54 54 0 45 68 59 50 28

150 57 51 57 60 0 78 69 63 39 12 6

113 109 105 91 101 89 0 31 42 50 26 24

3

0

49 47 29 72 65 59 28 7 0

5

3 __ 121 123 114 98 107 97 39 0

53 56 31 18 __

5

6

7

12 -

134 121 119 112 112 108 70 78 0 45 31 23

155 121 112 109 109 100 95 95 57

181 186 175 179 171 161 149 156 131 118 0 52

196 222 210 209 220 203 176 171 165 148 81 0 -

0

17 16

’Each entry represents the number of times that suture i (row)is in a more advanced state of closure that suture j (column).The rows and columns are ordered from earliest to latest closing sutures, with numbering of sutures following that given in Table 3.

intervals is disturbing, it is important to note that this is not a problem unique t o suture closure. Rather, all age indicators become progressively inefficient with advancing age due to the divergence between individuals with different biological aging rates. Biological aging rates may vary due to genetic and environmental factors such as nutrition. Because of the broad confidence intervals associated with our analyses, we must conclude that cranial sutures (endocasts) are not very useful for precisely aging older macaques. However, cranial suture closure can be used with caution as an age estimator in macaques. Rawlins (1975) investigated age-related changes in the pubic symphysis of subadult and adult macaques using a subset of the sample employed here. He found symphyseal morphology to be of little use in aging adult females because of extensive remodeling accompanying parturition. Limited estimates of age could be made in adult males using the extent of symphyseal fusion. However, Rawlins (1975) noted the serious need

for alternative aging methods €or adult primates. While endocranial sutural fusion cannot, by itself, provide very precise age estimates for adult macaque skulls (see Figs. 2, 3), it is the best currently available method. From a comparison of Meindl and Lovejoy’s (1985) Table 3 with our Table 5 , it appears that human endosutures begin closing at ages that are five or more years younger than the average ages a t which their ectocranial portions are reported to still be open. Thus cranial sutures of humans may close from the inside out, as suggested by Acsadi and Nemeskeri (1970:116):“ossification of sutures begins inwardly and spreads outwardly.” Meindle and Lovejoy (1985) claim that ectocranial sutures are superior to endocranial sutures for estimating age in humans because the mean deviations they obtained in ectocranial scoring of combined vault sutures for various ages are approximately half of the mean deviations found by Acsadi and Nemeskeri (1970) for their combined endo-

ENDOSUTURES OF MACAQUES

cranial vault sutures. However, the former authors combined seven small lengths of suture (“sites”)from the lambdoid, sagittal, and coronal sutures in their vault calculations whereas the latter workers used the overall degree of suture closure for coronal, sagittal, and lambdoid sutures (i.e., three variables) in their calculations. Thus, it is not clear that the variables used to represent vault sutures in the two studies differ only in regard to their externallinternal locations on the skull. Hence, mean deviations from these two studies do not appear to be comparable and therefore should not be used to draw conclusions about which type of suture is the best predictor of age. As shown in Table 6, endosutures of humans may be more sensitive than ectosutures for detecting details of the initiation of suture closure. That is, ectosutures that are beginning to close reflect the closing sequence of endosutures, but fail t o reveal the actual sequence in which cranial sutures begin to close k e . , the initiation sequence for endosutures). For instance, if one relied only on ectosutures, the important role of the sphenofrontal suture in human cranial growth would not be detected. The pattern of endosuture closure in the macaque can be compared to that for humans to determine whether a general primate pattern exists (Table 6). There are some broad similarities in pattern between the two species, with macaques and humans being more similar in the sequences in which their cranial sutures achieve closure, than they are in the sequences in which their sutures begin to close. For example, the basilar suture is the earliest closing suture for both humans (Krogman, 1978) and macaques, while the masto-occipital and rostral and caudal squamosal sutures (and to an extent, the sphenotemporal) achieve closure quite late in both species. Humans and macaques differ dramatically in their schedules for the sphenofrontal endosuture. The sphenofrontal suture of humans is the first (of those listed in Table 6) to begin closing, but the fourth to achieve closure. This accounts for the fact that it takes over ten times as long to close in humans as it does to close in rhesus monkeys (Table 5). It is tempting to speculate that this difference may be related to expansion of the frontal lobes and/or temporal poles in Homo relative to Macaca. Although the squamosal endosuture is the last t o close in both humans and macaques, its rostral portion begins to close early only

427

in the latter species. This difference is difficult t o interpret. It may simply be that the rostral squamosal is “riding along” with the sphenotemporal (its neighbor) in the initiation sequence of macaques. It is interesting to note that sulci from occipital regions of the cerebral cortex generally do not reproduce well on endocasts from macaques (or other monkeys) (Falk, 1978). Poor reproducibility of occipital sulci may be related to the late initiation and closure schedules for lambdoid, masto-occipital, and caudal squamosal sutures for macaques. (For reasons that are not entirely clear, endocasts prepared from human skulls reproduce overall poorer sulcal patterns than do endocasts from monkey skulls.) Our finding that rhesus monkeys lack sexual dimorphism in timing of suture closure mirrors numerous reports concerning cranial sutures of humans (Todd and Lyon, 1924; Meindl and Lovejoy, 1985; Acsadi and Nemeskeri, 1970). It may be that this is the general primate pattern, or that sexual dimorphism in closure of some sutures is too subtle to have been detected in this study. It is important to note that each cast was scored at one sitting. Consequently, the right and left side observations were not made independently, with the result that error may have been introduced to the determination of fluctuating asymmetry for this sample. The proportion of endocasts showing asymmetrical closure is generally lower than the figures reported by Zivanovic (1983) for humans. It is, however, difficult to interpret these results, as Zivanovic does not state whether bilateral sutures were scored independently within his sample. Additionally, proportion of asymmetrical cases is not a very telling measure, as it does not take into account the ordinal nature of suture closure. For example, a suture scored as 0 on the left and 3 on the right is considered in the same category with a suture scored 0 on the left and 1 on the right by Zivanovic (19831, but not here. With this in mind, one ofthe most interesting findings of this study is that there is a small but statistically significant directional asymmetry in closure of the masto-occipital and rostral squamosal sutures of macaques, with right sides preceding the left. (Asymmetry favoring closure of the right lateral lambdoid suture also approaches significance at the 0.05 level.) Although these sutures do display significant asymmetry in closure, we still chose to collapse all bilateral suture scores into single sutures. We sug-

428

D. FALK ET AL.

gest, however, that both sides of a cranium should be scored to determine whether the sides are symmetrical. Because the caudal squamosal does not begin to close in rhesus monkeys until after 15 years (Table 51, we are unable to determine whether or not closure in this portion of the squamosal is also asymmetrical. This would not be surprising however, since the caudal squamosal forms a complex with the other three sutures. Asymmetry in premature closure of the lambdoid suture occurs infrequently as a clinical condition (lambdoid craniosynostosis) in humans (Furuya et al., 1984; Hinton et al., 1984; Muakkassa et al., 1984). Interestingly, more cases of lambdoid craniosynostosis involve early fusion of the right than the left lambdoid suture and are associated with “flattening of the occipital bone on the affected side, compensatory bulging of the anterior ipsilateral portion of the calvaria.. . .” (Furuya et al., 1984:63). Thus, shape features that characterize the more prevalent right lambdoid craniosynostosis are associated with right frontal and left occipital projections (relative t o their counterparts). Less dramatic and nonpathological braidskull asymmetries called “petalias” occur in similar directions in human populations and are statistically associated with right-handedness (LeMay, 1977; Galaburda et al., 1978; LeMay et al., 1982). Shape assymetries that are less striking but in a similar direction to the right frontal and left occipital petalias that are modal for human skulldbrains have been documented for endocasts of Old World monkeys (LeMay et al., 1982). Although we are not suggesting that the asymmetries we report for suture closure in macaques are examples of craniosynostosis, our findings do suggest the possibility that cranial petalias of primates (including humans) may occur in conjunction with asymmetrical suture closure. If so, asymmetries of the skull may be due to differences in timing of developmental events on the two sides. ACKNOWLEDGMENTS

We thank Jim Neely for commenting on the manuscript, and the University of Puerto Rico is acknowledged for providing free access to the Cay0 Santiago skeletal collection. This research is supported by Public Health Service grant 7 R01 NS24904. LITERATURE CITED Acsadi G, and Nemeskeri J (1970)History of Human Life

Span and Mortality. Budapest: Hungarian Academic Society. Agresti A (1983) Testing marginal homogeneityfor ordinal categorical variables. Biometrics 39:505-510. Agresti A (1984) Analysis of Ordinal Categorical Data. New York John Wiley and Sons. Brooks ST (1955) Skeletal age at death: The reliability of cranial and pubic age indicators. Am. J. Phys. Anthropol. 13:567-589. Cheverud J (1981) Epiphyseal fusion and dental eruption in Macaca mulatta. Am. J . Phys. Anthropol. 56:157-168. Efron B (1982) The Jackknife, the Bootstrap and Other Resampling Plans. Philadelphia: The Society for Industrial and Applied Mathematics. Falk D (1978) External neuroanatomy of Old World monkeys (Cercopithecoidea). Contrib. Primatol. 15:l-95. Furuya Y, Edwards MSB, Alpers CE, Tress BM, Norman D, and Ousterhout DK (1984) Computerized tomography of cranial sutures. J. Neurosurg. 61:59-70. Galaburda AM, LeMay M, Kemper TL, and Geschwind N (1978) Right-left asymmetries in the brain. Science 199:852-856. Goodman LA, and Kruskal WH (1954) Measures of association for cross-classification.J. Am. Statist. AsSOC. 49:73%764. Hinton DR, Becker LE, Muakkassa KF, and Hoffman HJ (1984) Lambdoid synostosis Part 1: The lambdoid suture: Normal development and pathology of “synostosis.” J. Neurosurg. 61:333-339. Krogman W M (1978) The Human Skeleton in Forensic Medicine. Springfield, IL: Charles C. Thomas. LeMay M (1977) Asymmetries of the skull and handedness. Phrenology revisited. J. Neurol. Sci. 32:243-253. LeMay M, Billig MS, and Geschwind N (1982) Asymmetries of the brains and skulls of nonhuman primates. In E Armstrong and D Falk (eds.): Primate Brain Evolution: Methods and Concepts. New York: Plenum, pp. 263-271. McKern SW, and Stewart TD (1957) Skeletal Age Changes in Young American Males: Analysed from the Standpoint of Age Identification. Natick, MA: Headquarters Q.M. Res. and Dev. Command, Tech. Rep., EP-45. Meindl RS, and Lovejoy CO (1985) Ectocranial suture closure: A revised method for the determination of skeletal age a t death based on the lateral-anterior sutures. Am. J. Phys. Anthropol. 68:5746. Muakkassa KF, Hoffman HJ, Hinton DR, Hendrick EB, Humphreys RP, and Ash J (1984) Lambdoid synostosis Part 2: Review of cases managed at The Hospital for Sick Children, 1972-1982. J. Neurosurg. 61:340-347. Radinsky LB (1968) A new approach to mammalian cranial analysis, illustrated by examples of prosimian primates. J . Morphol. 124:167-180. Rawlins R (1975) Age changes in the pubic symphysis of Macaca mulatta. Am. J. Phys. Anthropol. 42:477-488. Sade DS, Chepko-Sade BD, Schneider JM, Roberts FS, and Richtsmeier J T (1985) Basic Demographic Observations on Free-Ranging Rhesus Monkeys. New Haven, CT: Human Relations Area Files. Singer R (1953) Estimation of age from cranial suture closure. J. Forensic Med. 1:52-59. Todd TW, and Lyon J r DW (1924) Endocranial suture closure. Part I. Adult males of white stock. Am. J. Phys. Anthropol. 7:325-384. Zivanovic S (1983) A note on the effect of asymmetry in suture closure in mature human skulls. Am. J. Phys. Anthropol. 60:431435.

All in-text references underlined in blue are linked to publications on ResearchGate, letting you access and read them immediate