close

Вход

Забыли?

вход по аккаунту

?

Endocranial suture closure in rhesus macaques (Macaca mulatta).

код для вставкиСкачать
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
i<k j<l
D= C
C nij n k l
(2)
i<k j>l
where nij and nkl are the cell frequency
counts for row i crossed with column j and
row k crossed with column 1, respectively.
Note that gamma ignores ties. For example,
an individual scored as 1,2 will not be
counted in the numerator or denominator of
gamma if compared to an individual scored
as 1,3,because these individuals are tied on
the first suture. The significance of gamma
can be assessed by dividing C - D by its
standard error to produce an approximate
z-score. The standard error of C - D is given
as (Agresti 1984:180):
where pi+ and p+j are the marginal proportions across j and 1, respectively, and where n
is the total sample size.
To examine directional asymmetry we
used delta (A), a measure of marginal homogeneity across the contingency table that
preserves the ordinal character of suture
closure. (One tests for correlations between
i<j
i>j
(4)
and can be assessed for significance using a
z-statistic given in Agresti (1983, 1984:207).
Because this z-statistic is only asymptotically normally distributed, we also calculated a probability by randomly reassigning
left and right sides, calculating delta, and
building a Monte Carlo distribution by repeated trials (using 999 runs).
With the existence of only a limited
amount of fluctuating and directional asymmetry in suture closure (see Results), bilateral scorings were collapsed into single variables. Consequently, for each of the nine
bilateral sutures, we used the observable
side if the other side was unobservable, and
randomly selected a side for each individual
if both sides were observable. This produced
observations on 12 sutures per individual
which we analyzed for interrelationships between age and suture closure.
Since initial inspection of scatter plots of
age (in months) against suture closure scores
illustrated that age increases geometrically
with respect to suture closure, logarithmic
age was regressed on suture closure in order
to derive predictive equations for chronological age based on suture closure state. Logarithmic scale ages correct for the problem of
heteroscedasticity, where there is increasing
age variation across increasing suture closure states. Using stepwise regression we
eliminated the less informative sutures to
construct a more parsimonious regression
equation. Rather than report the stepwise
multiple regression of log age on suture closure, we chose to average the retained sutures (analogous to Meindl and Lovejoy’s
(1985) composite suture scores) and present
a simple regression of log age on average
suture closure. These results were tested
with a jackknife regression analysis (Efron,
1982). Because we are primarily interested
in presenting a method for aging older individuals (where dental eruption and epiphyseal union are not useful), we recalculated
the regression using only those animals who
were 6 years or older at time of death.
In addition to calculating regressions of
log age on suture closure, we examined the
420
D. FALK ET AL.
order in which sutures begin to close and
(separately) the order in which closure is
achieved. Our sequences for endocranial sutures of macaques were compared with those
for humans (Todd and Lyon, 19241,as well as
the initiation and achievement sequences of
ectosutures for humans (Meindl and Lovejoy, 1985).
We also examined order of suture closure
without separating initiation and closure sequences by forming a matrix that contained
counts for the number of times across all
endocasts where the closure score for suture
i exceeded the state for suture j. We then
ordered the rows and columns of the matrix
so that elements would increase off the main
diagonal in the upper triangle and decrease
off the diagonal in the lower triangle. This
ordering was determined both by visual inspection and by forming all possible permutations for various submatrices out of the 12
by 12 matrix. Additionally, at this time we
tested for sex effects (difference in male/
female intercepts) and sex by suture interac-
tions (differences in malejfemale slopes of
age against suture closure) to determine
whether there are sex differences in suture
closure timing. As sex was not found to be an
important factor, we have not included the
results of these tests here.
RESULTS
Table 1 contains the summary statistics
for righaeft correlations between bilateral
sutures. The correlations between the left
and right sides are extremely high, suggesting that there is relatively little fluctuating
asymmetry in suture closure. All correlations exceed 0.90 (note that gamma is constrained between - 1and + 1)and are highly
significant (P < 0.01).
Table 2 contains the summary statistics
for directional asymmetry of suture closure.
Negative delta values indicate that the right
side sutures are in a more advanced state of
closure than are the left side. Alternatively,
a positive asymmetry score indicates that
the left side suture is more advanced. The
TABLE 1. Left/right correlations for nine bilateral sutures'
Large sample
Suture
Medial coronal
Lateral coronal
Medial lambdoid
Lateral lambdoid
Masto-occipital
Sphenofrontal
Sphenotemporal
Rostral squamosal
Caudal squamosal
Gamma
N
0.9798
0.9439
0.9893
0.9609
0.9581
0.9817
0.9256
0.9617
0.9613
328
325
317
312
281
304
307
313
302
Proportion of asymmetry
z
12.22
8.42
17.40
14.90
13.11
10.28
8.48
11.53
12.73
0.1341
0.1661
0.1325
0.1891
0.2313
0.1381
0.1401
0.1725
0.1954
'Correlations are expressed as Goodman and Kruskal's gamma. The significanceof gamma (against a null of gamma equals zero) can be
assessed using a large sample z-test. As all of the lefthight correlations here are highly significant, the z-value rather than the
probability value is presented.Sample sizes vary from suture to suture due to missing data. The proportionof asymmetry is the number of
lefthight pairs in different states of closure out of the total number of observed pairs.
TABLE 2. Directional asymmetry as measured by delta (difference between concordances and discordances)'
Suture
Medial coronal
Lateral coronal
Medial lambdoid
Lateral lambdoid
Masto-occipital
Sphenofrontal
Sphenotemporal
Rostral squamosal
Caudal squamosal
delta
z
Probabilitv (>zP
Monte Carlo
Drobabilitv2
0.0196
0.0183
-0.0110
-0.0243
-0.0504
0.0198
0.0247
-0.0378
-0.0005
1.2372
0.9162
-0.7283
-1.6144
-2.6069
0.8186
1.2055
-2.0819
-0.0251
0.108
0.180
0.233
0.053
0.004
0.206
0.114
0.019
0.490
0.123
0.171
0.137
0.070
0.006
0.117
0.118
0.018
0.509
'The significance of delta is assessed by a large sample z-test (one-tailed)and confirmed with a Monte Carlo randomization test (999
trials). Sample sizes for each suture are as in Table 1.
'One-tailed.
ENDOSUTURES OF MACAQUES
large sample z-score test indicates that there
is a significant amount of directional asymmetry (at P < .05, one-tailed) for the mastooccipital and rostral squamosal sutures, although even for these sutures the level of
asymmetry is small. Closure of the lateral
lambdoid suture is nearly significantly
asymmetrical at the 0.05 level ( P = 0.053).
These findings are substantiated with the
Monte Carlo randomization test. For all
three sutures the right sides are in a more
advanced state of closure than are the left.
To summarize the progression of suture
closure with age, Table 3 provides the percentage of endocasts for each suture within
each state of closure across yearly age intervals up to age 10years. Percentages included
for ages 10-15 years (120-180 months) are
lumped in the entries for 120 months. All
sutures follow an age progression of closure,
with the speed of closure varying between
sutures. Additionally, it can be determined
from Table 3 that while some sutures always
reach obliteration in older animals (e.g., the
basilar suture), other sutures persist in even
the oldest age category (e.g., the caudal
squamosal).
Table 4 presents the regression coefficients and summary statistics for a regression of log age (in months) on all 12 sutures
for the 285 monkeys who had completely
observable sutures. The highly significant (P
< 0.001) multiple correlation of 0.894 indicates that there is a strong relationship between age and stages of suture closure. The
standard error of 0.376 for this regression is
on a log scale and thus increases with increasing age. Additionally, the confidence
interval for age is asymmetrical around the
estimated age. For example, at an estimated
age of one year (12 months), 8.2 months is
one standard error below the estimated age,
while 17.5 months is one standard error
above. At an estimated age of 10 years, 6.9
years is one standard error below the estimated age, while 14.6 years is one standard
error above. An additional point from Table 4
is that the sutures are not of uniform importance in age prediction. As the partial regression coefficients (and associated t-tests) indicate, some sutures add little to the
determination of age.
Although we could present the multiple
stepwise regression, it would be impossible
to graph the confidence interval. Likewise,
because the confidence interval must be calculated using the covariance matrix between
the sutures, it is unlikely that researchers
42 1
would be willing to make such calculations
on a case-by-case basis. In any event, the
differencebetween the R2sfrom the stepwise
multiple regression and from the simple regression of age on average suture closure
was trivial.
Figure 2 shows the exponential curve and
the 95% confidence intervals for prediction
of individual cases for the average of seven
sutures (segments), including the rostral
sagittal, caudal sagittal, lateral lambdoid,
basilar, sphenofrontal, rostral squamosal,
and caudal squamosal. The equation for Figure 2, based on 295 cases, is log, age = 0.992
(average suture closure) + 2.595. There is a
strong and highly significant relationship
between log, age and average suture closure
(R2= 0.791, with a standard error of the
estimate equal to 0.374 and an increasing
uncertainty in age prediction with advancing suture closure. This conclusion was supported by a jacknife regression analysis. The
latter analysis showed that the age estimates are generally unbiased.
For those animals who were six years or
older a t the time of death, stepwise regression was used to select five sutures to average, including the medial coronal, caudal
sagittal, basilar, sphenotemporal, and caudal squamosal sutures. This analysis, based
on 130 individuals, resulted in the equation:
log, age = 0.726 (average suture closure) +
3.423 (R2= 0.46; SE of estimate = 0.252).
Although as one would expect, there is less
heteroscedasticity in the data, the confidence interval remained quite broad (see
Fig. 3). Even so, estimates of older animals
should be made using this second equation
due to its smaller standard error of the estimates.
The durations of closure for specific endocranial sutures are listed for rhesus monkeys and humans in Table 5. The data for
rhesus monkeys were determined from Table 3; those for humans were taken from
Todd and Lyon (1924). It is noteworthy that
the sphenofrontal suture takes over ten
times as long to close in humans as it does in
rhesus monkeys. (It is difficult to compare
the masto-occipital because it is not known
when this suture reaches closure in
macaques.) There is a long duration of closure in the sphenofrontal suture of humans
(see Table 6). The rostral squamosal of humans, on the other hand, begins to close
much later in life than it does in rhesus
monkeys, in whom rostral and caudal portions have distinct schedules (Table 6).
422
D.FALK ET AL.
TABLE 3. Percentages of endocasts by age for each
suture scored in the four closure states'
Age
(months)
0
(%)
1
(%)
Medial coronal suture
0
45.0
40.0
12
19.4
58.3
24
4.5
72.7
36
6.1
63.6
48
0.0
36.6
60
0.0
12.9
72
0.0
7.7
84
0.0
5.3
96
0.0
0.0
108
0.0
0.0
120
0.0
0.0
180f
0.0
0.0
Lateral coronal suture
0
5.0
70.0
12
8.6
62.9
24
4.5
59.1
36
0.0
48.5
48
2.4
19.5
60
0.0
9.7
0.0
18.5
72
84
0.0
0.0
96
0.0
0.0
108
0.0
5.6
0.0
0.0
120
180f
0.0
0.0
Rostra1 sagittal suture
0
70.0
30.0
12
62.9
37.1
24
50.0
50.0
36
12.1
84.8
48
4.9
70.7
60
0.0
50.0
72
0.0
30.8
84
0.0
16.7
96
0.0
13.3
108
0.0
5.9
120
0.0
0.0
180f
0.0
0.0
Caudal sagittal suture
0
75.0
20.0
12
64.7
35.3
24
72.7
27.3
36
28.1
59.4
48
12.2
43.9
60
3.4
34.5
72
0.0
19.2
84
0.0
0.0
96
6.7
6.7
108
0.0
6.3
120
0.0
0.0
180f
0.0
0.0
Medial lambdoid suture
0
78.9
15.8
12
68.6
28.6
24
68.2
22.7
36
33.3
48.5
48
17.5
40.0
60
10.0
30.0
72
7.7
34.6
84
0.0
0.0
96
6.3
0.0
108
0.0
12.5
State
2
Age
(months)
(%)
15.0
22.2
22.7
30.3
63.4
87.1
88.5
94.7
100.0
94.1
87.0
43.5
TABLE 3. (continued)
0.0
0.0
0.0
0.0
0.0
0.0
3.8
0.0
0.0
5.9
13.0
56.5
20
36
22
33
41
31
26
19
15
17
46
23
~~
25.0
28.6
36.4
51.5
78.0
90.3
77.8
100.0
100.0
88.9
93.5
69.6
0.0
3.7
0.0
0.0
5.6
6.5
30.4
20
35
22
33
41
31
27
19
16
18
46
23
0.0
0.0
0.0
3.0
24.4
46.7
61.5
61.1
66.7
41.2
52.2
15.4
0.0
0.0
0.0
0.0
0.0
3.3
7.7
22.2
20.0
52.9
47.8
84.6
20
35
22
33
41
30
26
18
15
17
46
26
0.0
20
34
22
32
41
29
26
19
15
16
46
23
0.0
0.0
0.0
0.0
0.0
5.0
0.0
0.0
12.5
43.9
58.6
73.1
84.2
80.0
31.3
41.3
8.7
0.0
0.0
3.4
7.7
15.8
6.7
62.5
58.7
91.3
5.3
2.9
9.1
18.2
42.5
60.0
50.0
88.9
93.8
56.3
0.0
0.0
0.0
0.0
0.0
0.0
7.7
11.1
0.0
31.3
0.0
0.0
19
35
22
33
40
30
26
18
16
16
0
(%)
1
(96)
Medial lambdoid suture
120
0.0
4.3
180+
0.0
0.0
Lateral lambdoid suture
0
73.7
26.3
12
57.1
40.0
24
45.5
54.5
36
24.2
57.6
48
15.0
45.0
60
6.7
26.7
72
3.8
34.6
84
0.0
5.6
96
0.0
12.5
108
0.0
6.3
120
0.0
8.7
180f
0.0
4.3
Masto-occipital suture
0
78.9
21.1
12
57.6
42.4
24
63.6
36.4
36
46.9
40.6
48
30.8
56.4
60
28.6
53.6
72
15.4
65.4
84
5.6
22.2
96
20.0
40.0
108
0.0
37.5
120
0.0
31.1
180f
0.0
13.6
Basilar suture
0
56.3
25.0
12
54.5
39.4
24
68.2
22.7
36
54.5
33.3
48
24.4
34.1
60
6.9
13.8
72
8.3
25.0
84
0.0
5.6
96
0.0
0.0
108
0.0
0.0
120
0.0
0.0
180+
0.0
0.0
Sphenofrontal suture
0
35.0
55.0
12
16.7
69.4
24
9.1
63.6
36
6.3
43.8
48
2.4
36.6
60
0.0
6.5
72
0.0
16.0
84
0.0
5.3
96
0.0
6.7
0.0
0.0
108
120
0.0
0.0
180f
0.0
0.0
Sphenotemporal suture
0
5.3
84.2
12
0.0
65.7
24
0.0
63.6
36
0.0
45.5
48
0.0
43.9
60
0.0
6.7
State
2
(%)
43.5
39.1
52.2
60.9
46
23
0.0
2.9
0.0
18.2
40.0
66.7
57.7
83.3
81.3
75.0
76.1
60.9
0.0
0.0
0.0
0.0
0.0
0.0
3.8
11.1
6.3
18.8
15.2
34.8
19
35
22
33
40
30
26
18
16
16
46
23
0.0
0.0
0.0
12.5
12.8
17.9
19.2
72.2
40.0
62.5
64.4
72.7
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
4.4
13.6
19
33
22
32
39
28
26
18
15
16
45
22
18.8
6.1
9.1
12.1
34.1
41.4
12.5
22.2
6.7
0.0
0.0
0.0
0.0
0.0
0.0
7.3
37.9
54.2
72.2
93.3
100.0
100.0
100.0
16
33
22
33
41
29
24
18
15
17
46
19
10.0
...
13.9
27.3
50.0
61.0
93.5
84.0
94.7
93.3
100.0
100.0
95.7
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
4.3
20
36
22
32
41
31
25
19
15
17
45
23
10.5
34.3
36.4
54.5
56.1
93.3
0.0
0.0
19
0.0
35
0.0
22
0.0
33
0.0
41
0.0
30
(continued)
ENDOSUTURES OF MACAQUES
TABLE 3. Percentages of endocasts by age for each
suture scored in the four closure states' (continued)
Age
(months)
0
(W)
1
(W)
State
2
(W)
3
(%)
N
Sphenotemporal suture
72
0.0
23.1
84
0.0
11.1
96
0.0
6.3
108
0.0
0.0
120
0.0
2.3
180+
0.0
0.0
76.9
88.9
93.8
100.0
93.0
91.3
0.0
0.0
0.0
0.0
4.7
8.7
26
18
16
17
43
23
Rostral squamosal suture
0
36.8
63.2
8.6
77.1
12
24
0.0
81.8
36
0.0
57.6
48
0.0
29.3
60
0.0
22.6
72
0.0
16.0
84
0.0
10.5
96
0.0
0.0
108
0.0
0.0
120
0.0
10.9
180+
0.0
0.0
0.0
14.3
13.6
42.4
65.9
77.4
80.0
78.9
100.0
88.2
78.3
56.5
0.0
0.0
4.5
0.0
4.9
0.0
4.0
10.5
0.0
11.8
10.9
43.5
19
35
22
33
41
31
25
19
16
17
46
23
Caudal squamosal suture
0
84.2
15.8
12
82.9
17.1
31.8
24
68.2
36
33.3
63.6
48
25.6
66.7
60
13.8
72.4
72
12.0
76.0
0.0
55.6
84
56.3
96
6.3
108
64.7
0.0
120
2.2
50.0
180+
0.0
26.1
0.0
0.0
0.0
3.0
7.7
13.8
12.0
44.4
37.5
35.3
47.8
73.9
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
19
35
22
33
39
29
25
18
16
17
46
23
'0, open; 1, closing; 2, closed; 3, obliterated. N is the number of
casts in the age range for which the suture could be scored.
TABLE 4. Regression
Variable
Medial coronal
Lateral coronal
Rostral sagittal
Caudal sagittal
Medial lambdoid
Lateral lambdoid
Masto-occipital
Basilar
Sphenofrontal
Sphenotemporal
Rostral squamosal
Caudal squamosal
Constant
of log,
Table 6 presents sequences in which endocranial sutures initiate and achieve closure for macaques and humans (Todd and
Lyon, 1924). The sequences for macaques
were determined from Table 3: For each
suture, initiation of closure is the age when
0.0% of the endocasts are still open (0).The
end of closure is the age when 0.0% of the
endocasts are open (0)and 0.0% are in the
process of closing (11, i.e., the age at which all
endocasts are either closed (2) or obliterated
(3).Similar sequences are presented in Table
6 for ectosutures of humans (determined
from data presented in Table 3 of Meindl and
Lovejoy, 1985). Table 6 shows that endosutures of humans and macaques begin to close
in a different sequence from that in which
they achieve closure. Ectosutures of humans, on the other hand, both begin and
achieve closure in the same sequence as
endosutures of humans achieve closure. (We
should caution that the ectosuture sequences were determined from average ages
for suture scores (Meindl and Lovejoy, 19851,
a method that differs from those used for
endosutures. Nevertheless, the ectosuture
sequences were consistent across all four of
Meindl and Lovejoy's suture scores. We
therefore believe it is instructive to compare
the sequences presented in Table 6.)
Sutures each have their own schedule of
closure, speeding up and slowing down at
various stages (Todd and Lyon, 1924). This
accounts for differences in the sequences of
endosutures that are beginning to close and
those that have achieved closure for both
rhesus and humans (Table 6). To smooth
age in months on suture closure (scored as 0, 1,2, or 3) for all animals'
Partial
regression
coefficient
Partial
correlation
coefficient
t
Probability2
0.050
0.045
0.209
0.120
-0.013
0.087
0.001
0.104
0.187
0.003
0.178
0.099
2.426
0.040
0.029
0.249
0.153
-0.016
0.093
0.001
0.156
0.136
0.002
0.139
0.084
0.000
0.915
0.752
3.744
2.264
-0.277
1.702
0.025
3.278
3.393
0.053
3.422
1.952
23.274
0.361
0.453
0.000
0.024
0.782
0.090
0.980
0.001
0.001
0.957
0.001
0.052
0.000
' N = 285; multiple R = 0.894; F = 90.55,P < 0.001; Standard error of estimate = 0.376
Two-tailed.
423
424
D. FALK ET AL.
25
,
I
:
/ =
: / -i
i
O
,
09
’
,
’
,
1
05
1
1
I
I
1
10
i
I
I
15
l
l
’
~
20
l
l
l
~
~
25
Average S u t u r e Closure
Fig. 2. Age at death in years plotted against the average suture closure score over seven suturesfor the entire
Sam le of 295 individualsat all ages. Stars represent the predicted age at deathfor individual animals.The upper
and Power limits of the 95%confidence interval(basedon loglinear regression)are indicatedbythe center line and
the two surrounding lines.
over these “closure spurts,” our data for
rhesus monkeys are presented in a matrix
(Table7) that orders sutures across our sample according to their relative degree of closure, but without regard to their specific
ages for beginning and achieving closure.
Each entry represents the number of times
that suture i (row) is in a more advanced
state of closure than suture j (column).Ties
between sutures have been excluded. Rows
and columns in Table 7 are ordered by the
relative order of suture closure in the
macaque cranium. The order of closure is
basilar, followed by lateral coronal, sphenotemporal, medial coronal, rostral squamosal, sphenofrontal, caudal sagittal, rostral
sagittal, medial lambdoid, lateral lambdoid,
masto-occipital, and caudal squamosal. It
should be noted that the order of closure is
not always clear-cut. In particular, sphenotemporal, medial coronal, and rostral
squamosal all close at approximately the
same time, as is also true of the rostral and
caudal sagittal (thus the division into anterior and posterior portions may be unnecessary).
To measure the fit between our idealized
order of closure and the actual data, we may
compare the number of times that an “earlier” closing suture has a higher score than a
’later” closing suture with the number of
times that this pattern is reversed. The proportion of the first count out of the total gives
the percentage of accordance out of all counts
containing information on order of suture
closure. The observed percentage of 74.3%is
relatively high and indicates not only that
our ordering is correct, but also that the
pattern of suture closure is fair1 uniform
across cases. It should be note that the
general sequenceof closure revealed in Table
7 differs from that shown for sutures of
B
l
l
ENDOSUTURES OF MACAQUES
0
I
425
,
I
1
I
I
' 6
13
I
I
I
1
23
-4verage S u t u r e Closure
Fig. 3. Age at death in years plotted against average suture closure score for five selected sutures. Only
animals six years or older at the time of death are included (n = 130).Symbols as described in legend to Figure 2.
TABLE 5. Duration of endocranial suture closure'
Homo
Macaca
Sphenotemporal
Rostrosquamosal
Coronal
Sphenofrontal
Sagittal
Lambdoid
Masto-occipital
Caudosquamosal
Duration
(years)
1-15
2>15
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.
Документ
Категория
Без категории
Просмотров
3
Размер файла
998 Кб
Теги
closure, endocranial, macaque, rhesus, mulatta, macaca, suture
1/--страниц
Пожаловаться на содержимое документа