close

Вход

Забыли?

вход по аккаунту

?

Cranial deformation and nonmetric trait variation.

код для вставкиСкачать
AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 90:3548 (1993)
Cranial Deformation and Nonmetric Trait Variation
LYLE W. KONIGSBERG, LUCI A.P. KOHN, AND
JAMES M. CHEVERUD
Department of Anthropology, University of Tennessee, Knoxville,
Tennessee 37996 (L.W.K.); Department ofdnthropology, Field Museum of
Natural History, Chicago, Illinois 60605 (L.W.K.); Department of
Anatomy and Neurobiology, Washington University School of Medicine,
St. Louis, Missouri 63110(L.A.P.K., J.M.C.)
KEY WORDS
cans
Biological distance, Skeletal biology, Native Ameri-
ABSTRACT
Cranial deformation is known to influence many traditional
craniometric variables, but its effects on nonmetric trait variation are not well
characterized. In this study, we examine the effects of three types of deformation (annular, lambdoid flattening, and fronto-occipital) on nonmetric traits,
using a large sample of protohistoric and prehistoric crania. Our results
indicate that a few traits are increased or decreased in relative frequency by
particular types of deformation, but that these effects have little impact on the
calculation of biological distances between groups. o 1993 Wiley-Liss, Inc.
Artificial cranial deformation has obvious
and quantifiable effects on human cranial
morphometry (Blackwood and Danby, 1955;
Moss, 1958; Bjork and Bjork, 1964; McNeill
and Newton, 1965; Hellmuth, 1970; Anton,
1989; Kohn and Cheverud, 1991; Cheverud
et al., 1992). In contrast to these unambiguous effects on metric variation, influence of
deformation on cranial discrete or nonmetric trait frequencies has not been clearly resolved in the literature. Nearly 100 years
ago, Dorsey (1897) suggested that the high
relative frequency of coronal ossicles in a
sample of Kwakiutl crania at the Field Museum was caused by the annular deformation practiced by this group. In contrast,
Sullivan (1922) could find no clear association between deformation and discrete trait
frequencies for North American samples.
However, these two early studies did not
compare nonmetric trait frequencies between deformed crania within individual
populations, and consequently it was impossible to tell whether trait frequency differences were due to deformation or to between-population variation. In more recent
studies, researchers have contrasted trait
frequencies across deformation status (or
some proxy measure) within populations
0 1993 WILEY-LISS,
INC
(e.g., Bennett, 1965; Ossenberg, 1970; Buikstra, 1976; El-Najjar and Dawson, 1977;
Gottlieb, 1978; Shipman et al., 1990). Further, Pucciarelli (1974) has shown in experimental studies of rats that deformation can
increase the relative frequency of wormian
bones. Nonetheless, there is no current consensus on the importance of artificial cranial deformation in influencing nonmetric
trait frequencies among humans.
That cranial deformation could affect the
frequencies of nonmetric traits follows from
the empirical (and presumably developmental) relationship between metric and nonmetric cranial variation. Corruccini (19761,
Cheverud et al. (1979), and Richtsmeier et
al. (1984) have documented the interdependency of nonmetric and metric variation in
the human and rhesus cranium. Given their
empirical findings, as well as the presumption that there are environmental correlations among morphological aspects of the
cranium, it is possible that deformation
Address reprint requests to Lyle W. Konigsberg, Department
of Anthropology, 252 South Stadium Hall, University of Tennessee, Knoxville,Tennessee 37996-0720.
Received August 8,1991; accepted June 28,1992.
36
L.W. KONIGSBERG ET AL
could act jointly on both nonmetric and metric variation, at least a t a local level within
the cranium.
There are varied reasons why the effect of
cranial deformation on nonmetrics is unresolved. First, any simple contrast of “deformed versus ‘hot-deformed” does not take
into account the fact that there are different
types of deformation, nor the fact that intensity of deformation can vary widely. In this
study we consider the effects of three different types of deformation: annular, lambdoid
flattening, and fronto-occipital, in order to
examine whether they may have differential
effects on nonmetric trait frequencies. Additionally, we treat deformation a s a n ordinal
categorical variable so as to account for differing intensity of deformation. Second, the
classification of “deformed” versus “not-deformed” within past studies is often very
ambiguous, and is rarely supported by any
examination of repeatability of the scoring.
In contrast to the previous situation, we begin our examination of the effects of cranial
deformation by discussing a n intraobserver
repeatability study of deformation scoring.
Finally, the statistical methods for assessing the effects of deformation have not always been appropriate. For example, Buikstra (1976) criticized a n earlier study of
Ossenberg (1970) for not accounting for sideto-side correlation in bilateral traits. I n the
following analysis, we attempt to circumvent some of these previous statistical pitfalls by using more appropriate methods,
including both probit analysis and a multivariate distance measure which properly accounts for correlations between traits
(Blangero and Williams-Blangero, 1991).
MATERIALS AND METHODS
We scored deformation status, age, gender, and presencelabsence of 39 nonmetric
cranial traits (see Table 2 for a listing of
traits) using a sample of 447 crania from the
Hopi, Nootka, Kwakiutl, and a prehistoric
Peruvian series. The Hopi skeletal collection
is from the site of Old Walpi (also known as
Kuchaptavela, or Ash Hill). By association
with ceramics and ethnohistoric accounts,
the material from this site can be dated to
1300-1680 A.D. The skeletal material was
collected by C.L. Owen in 1901 and is main-
tained at the Field Museum of Natural History (see e.g., Dorsey and Voth, 1902; Martin
and Willis, 1940; and various Field Museum
Annual Reports for passing references to
Owen’s excavation). Although the collection
has been used in previous studies (e.g., ElNajjar, 1978; Cheverud et al., 1979) there is
no detailed archaeological account available. The Nootka and Kwakiutl skeletal collections derive from protohistoric or very
early historic groups, and were obtained primarily by Franz Boas, George Dorsey,
George Hunt, and C.F. Newcombe (see Cybulski, 1975; Cole, 1985 for sample and historical information, respectively). These collections are split between the Field Museum
and the American Museum of Natural History. The prehistoric Peruvian material is
from the necropolis at Ancon, and was collected by George Dorsey in 1892 (Dorsey,
1895). Ancon is a multicomponent site, so
the material a t Field Museum dates to a
broad time range. Artifactual associations
indicate that the material is from the Huari
Empire dating from 600 A.D. to 1450 A.D.
(Menzel, 1977).
The mode and form of deformation varies
substantially across these different collections. For the Hopi, children were cradleboarded and this often resulted in lambdoidal flattening (Dennis and Dennis, 1940).
The Kwakiutl and Nootka practiced annular
deformation, although the intensity of the
deformation varied between these two
groups (Cybulski, 1975) and probably across
regions within cultural groups a s well (Boas,
1921). Among the crania from Ancon, frontooccipital deformation is the most common
form of cranial modification (Reichlen,
1982).
All material at the Field Museum (380
crania) was scored by Konigsberg and Cheverud, while the material at the American
Museum (67 crania) was scored by Kohn.
Cranial deformation was scored on a n ordinal scale a s “not deformed (n.d.), “slightly
deformed” (s.d.1, “deformed (d.), or “much
deformed (m.d.1. We use “not deformed to
refer to any cranium where there is no indication that artificial deformation has reshaped the cranial vault. “Slightly deformed refers to crania where there is some
indication t h a t the vault has been reshaped,
CRANIAL. DEFORMATION AND NONMETRIC TRAITS
but where the deformation might be overlooked if we did not know the provenience
(and hence the culturally associated deformation practices) for the skull. We use “deformed for crania where there is no question but that the cranial vault has been
artificially reshaped, and “much deformed”
for crania where the reshaping is extreme.
In addition to the deformation scoring for
the Field Museum material by Konigsberg
and Cheverud made in 1989, there was also
an independent scoring of deformation made
by Konigsberg between 1985 and 1987 while
he was conducting an inventory of the Field
Museum skeletal collection. These two independent ordinal categorical scorings will be
used to determine the repeatability of the
deformation classifications.
We measure repeatability by finding the
polychoric correlation between the 19851987 and the 1989 scorings of deformation.
Other models of repeatability are available
for ordinal categorical data (see Agresti,
1988), but we use polychoric correlation because it allows for an underlying normal distribution of deformation status. Polychoric
correlation is the multiple-class analog of
the more commonly known tetrachoric correlation (Olsson, 1979). The tetrachoric correlation measures the underlying bivariate
correlation on which thresholds are placed
parallel to both the ordinate and abscissa to
produce four probability density regions.
The densities in these regions give the expected proportions of cases for a 2 x 2 contingency table. Thus, a tetrachoric correlation expresses the correlation between two
binary traits.
A polychoric correlation extends this idea
by allowing more than one threshold parallel to the ordinate andor the abscissa (see
Fig. 1). With four classes of deformation
(n.d., s.d., d., and m.d.1 the cross-tabulation
of the 1989 against the 1985-1987 scorings
produces a 4 x 4 contingency table. Assuming that cranial deformation is a normally
distributed variable which we observe as 4
ordered classes, maximum likelihood estimation can be used to find the standard bivariate normal distribution describing the
relationship between the two scorings. The
model can be parameterized using three
thresholds for the 1985-1987 scoring, three
37
Comparison of 1985 and 1989
Scoring of Deformation
-3
-2
I
I
I
I
I
-1
0
1
2
3
1985 Scoring
Fig. 1. Graphic representation of polychoric correlation between 1985-1987 and 1989 scorings of cranial
deformation. Axes are labelled in standard deviation
units, and there are three thresholds in each direction
cutting the graph into 16 regions. The bivariate correlation (r = 0.891 is shown with isolines at every 0.02 units.
thresholds for the 1989 scoring, and a Pearson correlation. Equality of the 1985-1987
and 1989 thresholds is examined to determine whether there was any change in the
scoring procedure across time. In other
words, this test determines whether there
was any change in the frequency of crania
assigned to each of the deformation classes
in 1985-1987 as versus 1989. If the thresholds are equal, then the polychoric correlation can be used as a measure of intraobserver repeatability for the deformation
scoring. The polychoric correlation and
threshold values were estimated using the
FORTRAN function BIVNOR (Baughman,
1988) and GEMINI (Lalouel, 1979) to maximize the log-likelihood function.
Because the underlying continuous distribution for the scoring of deformation suggested that the mean within-group deformation scores were approximately linearly
arranged (see Results), we coded the deformations as “not deformed” equal 0; “slightly
deformed equal 1; “deformed”equal 2; and
“much deformed equal 3. Age was coded as
1 (less than 16 years); 2 (16 to 30 years); 3
(30 to 50 years); and 4 (over 50 years), while
gender was coded as 1 (male); 2 (female); or
38
L.W. KONIGSBERG ET AL
likelihood is asymptotically distributed as a
chi-square statistic with 1 degree of freedom. Because different forms of deformation
are represented within this sample, we also
test whether the deformation effects are
equal across samples by reparametrizing
equation 1 to include deformation-by-population interaction. This saturated model (see
Agresti and Agresti, 1979 for a definition of
this concept) then contains 6 parameters:
age and sex effects which are assumed to be
equal across the four populations and deformation effects within each of the four populations. The restricted model contains the
age effect, sex effect, and a common deformation effect across all populations. The loglikelihood ratio test comparing these two
models has three degrees of freedom. All
probit log-likelihood functions were maximized using GEMINI (Lalouel, 1979), with
the normal integrals evaluated using the
FORTRAN function ALNORM (Hill, 1973).
To examine the effect of deformation on
the calculation of biological distances we calculated the morphological distances between the Hopi, Nootka, Ancon, and Kwakiutl samples separately for the nondeformed
sample (n.d.1 and the deformed sample (s.d.,
d., or m.d.1. The distance measure we used is
an extension of Mahalanobis’ D2 to cover discrete traits caused by thresholds on polygenwhere P(t = 1)is the probability that an in- ically determined liabilities (Blangero and
dividual will display a particular nonmetric Williams-Blangero, 1991). The distance betrait, +[f(x)] is the standardized normal in- tween two samples can be written as:
tegral from negative infinity to f(x), ci is a
constant, and the beta weights are regressions on sex, age, and deformation status.
The sum of the product of the observed
counts (number of individuals in a particu- where (z, z2) is a column vector of differlar classification of age, sex, deformation ences between t threshold values for two
status, and trait presencelabsence) with the samples, and R is a matrix of pooled withinnatural log of P(t = 1)or P(t = 0 ) is propor- group tetrachoric correlations between
traits. The thresholds for the z vectors were
tional to the log-likelihood.
Given the current method of coding, a pos- estimated using the probit regression shown
itive beta weight for deformation indicates in equation 1, which assumes homogeneous
that the trait frequency increases with de- age and sex effects across samples. In this
formation, while a negative weight indicates case, ci becomes a vector of threshold values
that the trait frequency decreases with de- (i.e., z) within populations after controlling
formation. To test whether these regres- for common age and sex effects. The tetrasions are significantly different from zero, choric correlations were estimated using biwe reestimate the probit parameters above variate probit analyses (see Ashford and
with the deformation effect set equal to zero. Sowden, 1970) with separate age and sex
Negative two times the difference in the log- effects for each pair of traits. The bivariate
9 (unknown). The nonmetric cranial traits
were scored as 0 (absent), 1 (present), or 9
(unobservable). Seventeen bilateral traits
and 5 midline traits were scored for a total of
39 traits.
To determine whether cranial deformation has an effect on the occurrence of individual cranial nonmetric traits we used
univariate probit analyses. For these analyses we disregard the fact that many of the
traits are intercorrelated, and that there are
often strong correlations between left and
right sides for the bilaterally scored traits.
In a previous study (Konigsberg et al., 1991)
we used bivariate probit analyses to control
for leftlright correlations in bilateral traits,
but given the complexity of interpreting the
results from these analyses, we have not included them here. We solely use the univariate probit analyses to look at individual effects on relative trait frequencies, and do not
try to use these analyses to make an omnibus statistical statement (which is reserved
for the following multivariate analyses).
For the univariate probit analyses, the
probit takes the following form:
~
CRANIAL DEFORMATION AND NONMETRIC TRAITS
39
TABLE 1 Repeatabclrty of cranral deformation scoring
Observed 1985-1987 and 1989 scorings
1989 scoring
nd
nd
sd
d
md
1985-1987 scoring
sd
134
20
2
0
30
51
29
0
Expected values, separate 1985-1987 and 1989 thresholds (polychoric correlation = 0.8890):
1989 scoring
nd
sd
md
0
0
4
8
d
md
~~
~~
1985-1987 scoring
d
4
21
76
1
nd
sd
d
md
134 28
20 42
2 27
0 00
32 01
49 82
26 15
0 01
Expected values, equal 1985-1987 and 1989 thresholds (polychonc correlation = 0 8868)
1989 scoring
n.d.
s.d.
n.d.
134.32
26.02
26.02
50.13
s.d
1985-1987 scoring
23.07
2.56
d.
m.d.
0.00
0.01
~~
2 67
19 88
78 32
2 70
0 00
0 01
5 99
5 47
d.
2.56
23.07
78.15
4.22
m.d.
0.00
0.01
4.22
5.63
-
0.1520 for the slightly deformed, 1.0294 for
the deformed, and 2.3284 for the much deformed classification (see Kelley, 1924 for
the method of converting threshold values to
RESULTS
within-class means). Because these 4
Table 1 contains the observed cross-tabu- within-class means are remarkably close to
lation of the 1985-1987 and 1989 scorings of being on a linear scale, we simplify the
cranial deformation for the Field Museum univariate probit analysis by coding deforskeletal material. In addition, this table con- mation as 0 (n.d.1, 1 (s.d.1, 2 (d.), and 3
tains the expected frequencies from the (m.d.).
Table 2 contains the deformation beta
polychoric correlation analysis with unequal
thresholds across the two marginal distribu- weights for each sample from the univariate
tions, and the analysis with equal marginal probit analyses. Although the age and sex
thresholds. The log-likelihood ratio xz for beta weights are not shown here, these pacomparison of these two models (unequal rameters were also estimated within each
versus equal marginal thresholds) is 4.6641, sample when possible. From a cursory exwith 3 degrees of freedom (P = 0.1981). This amination of Table 2, it is apparent that denonsignificant result indicates that the formation has a significant effect on the ocmarginal distributions for 1985-1987 and currence of only a minority of traits.
1989 are equal, so that there was no “tempo- Specifically, in the Hopi, deformation acts to
ral drift” in the deformation scoring proce- decrease the relative frequency of left
dure. Additionally, the high positive poly- masto-occipital ossicles, and increases the
choric correlation (r = 0.8868, P < 0.0001) relative frequency of the right foramen spiindicates the considerable repeatability of nosum open, right foramen of Huschke, and
sagittal ossicles. Among the Nootka, deforscoring for deformation.
The three thresholds from Table 1 were mation increases the relative frequency of
estimated at -0.1796, 0.4955, and 1.9439. left and right coronal ossicles, while for AnThese thresholds correspond to deviations of con, deformation acts to increase the relathe class means from the grand mean equal tive frequency of sagittal ossicles. Among
to -0.9157 for the not deformed class, the Kwakiutl, deformation decreases the
integrals for the bivariate probit analyses
were evaluated using the FORTRAN function BIVNOR (Baughman, 1988).
40
L.W. KONIGSBERG ET AL.
TABLE 2. Probit remession coefficients of trait fresuencres on deformation status (left sides listed before right)
Trait
Epipteric bone
Asterionic bone
Parietal notch hone
Larnhdoid 0s.
Masto-occipital 0s.
Coronal 0s
Infraorhital suture
Supraorbital foramen
Access. infraorb. f.
Divided hypoglossal
Postcondylar canal
F. wale open
F. spinosurn open
F. Huschke
Mastoid f. exsutural
Ohelionic foramen
Access. less. palat.
Metopic suture
Bregmatic bone
Apical hone
Inca bone
Sagittal hone
I
Houi
Nootka
Ancon
Kwakiutl
-0.2343
0.0205
-0.0794
0.0105
-0.4054
-0.2444
0.3266
0.1275
-0.4055*
-0.1868
0.5339'
0.3420
-0.0437
0.3420
-0.0437
0.1302
0.2438
0.0327
0.0149
- 0.4663
-0,2582
0.2623
0.2398
0.1160
0.01861,2
1.0282l*
0.1416
0.5262*
0.2350
0.1119
-0.1803
0.0386
-0.0645
-0.0104
0.1170
0.0766
0.0108'
0.5996
0.2077
0.1652
0.2985
0.0406
-0.1187
-0.0609
-0.0868
0.2507l
0.1678
0.6001"
0.9588
-0.2503
0.2658
-0.3729
-0.2331
-0.1640
-0.0897
0.2038
-0.0348
-0.0559
0.3589
-0.2253
-0.2898
0.0367
-0.1259
0.0395
0.1528
0.2716
0.0573
0.1333
-0.0755
0.2191
0.1047
-0.4366
0.0971
0.7214l
-0.6904*
-0.3731
0.1107
0.1016
0.1539
0.2731*
0.0815
-0.0477
0.2841
0.5735*
0.8008+
0.5191*
0.0531
-0.0026
-0.0067
-0.1422
-0.0613
-0.1082
-0.0751
0.0138
-0.0515
0.0630
0.1628
0.3611
0.0279l
-0.2703
-0.1401
-0.2614
0.2462
0.0849
-0.0042
0.1015
-0.0577
0.0335
-
____
0.7200"
-
____
____
-0.1433
-0.2013
0.2413
0.4065
-0.0009
0.1547
-0.2617
-0.1300
-
__--
-0.0110
-
_-__
-0.0110
-0.0906
-0.1235
-0.0893
-0.0579
0.0917
-0.0006
-0.1166
0.1435
-0.1938
0.0308
-
-
__--
0.0602
-0.2071
0.2028
0.2026
-0.1714
0.0546
-0.0095
0.1573
0.0766
0.2172
0.6566'.'
0.71811,'
0.4279*
-
____
-___
-0.1142
-0.1121
-0.0769
Sex effect could not be estimated.
'Age effect could not be estimated.
*Significantly different from zero at P < 0.05(two-tailed)
relative frequency of left epipteric bones,
while increasing the relative frequency of
right parietal notch bone, right masto-occipital ossicles, and left and right coronal ossicles. These changes in trait frequencies can
be observed in Appendix Figure A, which
contains the trait frequencies tabulated
across undeformed crania (n.d.1 and deformed crania (s.d., d., and m.d.1. The sample sizes in Appendix Figure A are slightly
larger than those used for the probit analyses, because crania of unknown sex were
eliminated from the probit analyses.
Aside from the question of individual effects of deformation within populations,
there is also the question of population-bydeformation interaction effects on the occur-
rence of nonmetric traits. Because there are
different forms of deformation across the
populations (occipital flattening in the Hopi,
fronto-occipital deformation for Ancon, and
annular deformation among the Nootka and
Kwakiutl), any interaction of deformation
with population indicates differential effects
of these various forms of deformation on
trait presence. Table 3 lists the probability
values from likelihood ratio chi-square tests
comparing three different models. The first
model is saturated, and includes a separate
deformation-within-population effect for
each population. The second model is a homogeneous effects model, in which there is
one deformation parameter across populations. In other words, the homogeneous
CRANIAL DEFORMATION AND NONMETRIC TRAITS
41
TABLE 3. P-values from comparisons of saturated to homogeneous deformation effects model, saturated to no
deformation effects model, and homogeneous effects to no effects model
Trait
Epipteric bone
Asterionic bone
Parietal notch bone
Lambdoid 0s.
Masto-occipital 0s.
Coronal 0s.
(-AP
Infraorbital suture
(-A)
Supraorbital foramen
Access. infraorb. f.
Divided hypoglossal
Postcondylar canal
F. ovale open
F. spinosum open
F. Huschke (-N)
(-N)
Mastoid f. exsutural
Obelionic foramen
Access. less. palat.
Metopic suture
Bregmatic bone
Apical bone
Inca bone (-H, -N)
Sagittal 0s. (-A)
Saturated
vs. homog.
Saturated
vs. no deform.
Homog. vs.
no deform.
0.0477"
0.6163
0.4382
0.8307
0.4209
0.1276
0.5003
0.7404
0.0239"
0.0440*
0.8547
0.5296
0.8354
0.5660
0.7026
0.5557
0.3148
0.9117
0.9629
0.2980
0.8280
0.3131
0.3662
0.0996
0.9985
0.0267*
0.5864
0.0195"
0.4834
0.6008
0.3756
0.5482
0.9928
0.8128
0.6532
0.5522
0.4967
0.1556
0.0583
0.0916
0.7232
0.3369
0.6500
0.4579
0.2046
0.1635
0.8426
0.0500"
0.0555
0.00011'
0.0214*
0.9092
0.6500
0.7530
0.5076
0.4523
0.8774
0.9793
0.4183
0.8218
0.3670
0.4903
0.1572
0.9998
0.0505
0.7676
0.0332"
0.1015
0.5205
0.5159
0.3152
0.9824
0.7620
0.7930
0.5191
0.6252
0.3641
0.0001"
0.7758
0.6001
0.1754
0.2071
0.3661
0.6267
0.0415"
0.6924
0.8303
0.2866
<0.0001"
0.0002*
0.7032
0.4977
0.4822
0.2680
0.7243
0.4125
0.6953
0.6324
0.4244
0.3896
0.6173
0.5474
0.9395
0.6082
0.7890
0.3569
0.0215*
0.2432
0.6982
0.1055
0.5780
0.3414
0.8065
0.2861
0.6348
0.9522
<0.0001*
*Significant difference between two models a t P < 0.05
'-A, -H, and -N indicate that Ancon, Hopi, and Nootka were deleted.
model is identical to the saturated model,
save that the deformation-within-population effects are assumed to be equal across
all populations. Because these effects are assumed to be equal k e . , homogeneous across
populations), they can be replaced with a
single deformation effect parameter. The final model of no deformation effects fixes the
deformation parameter to zero. In other
words, the final model allows for different
trait frequencies across populations, but assumes that there is no effect of deformation
on the trait frequencies. For some analyses,
the deformation effect could not be estimated within a few populations, and consequently had to be assumed to equal zero.
This was true for Ancon within the right
coronal ossicle, right infraorbital suture,
and sagittal ossicle analyses, for the Nootka
within the left and right foramen of Huschke
analyses, and for the Hopi and Nootka
within the inca bone analysis.
Table 3 indicates that there are heterogeneous effects of deformation across the left
epipteric bone, left and right masto-occipital
ossicles, right foramen spinosum open, and
right foramen of Huschke traits. Additionally, there are homogenous effects of deformation on the left lambdoid ossicles, left and
right coronal ossicles, the right mastoid foramen exsutural, and the sagittal ossicles.
In general, all effects represent increases of
wormian bones with deformation, except for
the case of left epipteric bones in the Kwak-
42
L.W. KONIGSBERG ET AL.
TABLE 4. Biological distance matrix between the Hopi,
Nootka, Ancon, and Kwakiutl, with distances among the
deformed series above the diagonal, and distances among
th.a undeformed series below the diagonal
Hopi
Nootka
Ancon
Kwakiutl
0,0000
0.7794
0.7427
1.5920
0.6751
0.0000
0.3147
0.5462
0.5174
0.3135
0.0000
0.7246
1.3256
0.6756
0.9725
0.0000
~
Hopi
Nootka
Ancon
Kwakiutl
~~
iutl. For this latter case, deformation decreases the relative frequency of left epipteric bones. Only three foramina1 traits are
affected by deformation. Right foramen spinosum open and right foramen of Huschke
are both increased by deformation, but this
effect is only present in the Hopi. Finally,
the relative frequency of the occurrence of
an exsutural left mastoid foramen is increased by all forms of deformation.
Given that the univariate effects of deformation on nonmetric trait occurrence do not
appear to be substantial, we need to consider the multivariate effects as represented
on the calculations of biological distances
between groups. Table 4 contains the
threshold-based Mahalanobis distances divided by the number of traits in the calculation. The distances are shown separately for
the nondeformed (n.d.1 and deformed (s.d.,
d., and m.d.1 series. For these calculations,
sides were randomly selected within crania
for bilateral traits, and traits were eliminated from the calculations if they were invariant within a sample. Random side selection provides the benefit of not making the
relative trait frequency dependent on preservation (see Green et al., 1979; Korey,
19801, and does not limit the sample to only
those crania where both sides are observable (see Konigsberg, 1987 for a further discussion).
These eliminations resulted in 15 traits
being used €or the nondeformed series (asterionic bone, parietal notch bone, lambdoid
os., masto-occipital os., infraorbital suture,
supraorbital foramen, accessory infraorbital
foramen, divided hypoglossal canal, postcondylar canal, f. ovale open, f. Huschke,
mastoid f. exsutural, obelionic f., accessory
lesser palatine f., and sagittal 0s.) and 16
traits for the deformed series (epipteric
bone, asterionic bone, parietal notch bone,
lambdoid os., masto-occipital os., coronal os.,
infraorbital suture, supraorbital foramen,
accessory infraorbital foramen, divided hypoglossal canal, postcondylar canal, f. ovale
open, mastoid f. exsutural, obelionic f., accessory lesser palatine f., metopic suture).
The distances shown in Table 4 are also represented in Figure 2, which is a principal
coordinates plot of the distances. Because
there are only four samples, the three-dimensional plots in Figure 2 completely represent the distance information shown in
Table 4. The minimum spanning trees are
also depicted in Figure 2. It is clear from an
examination of Table 4 and Figure 2 that
cranial deformation has not markedly altered the between-population relationships.
While the Hopi sample is slightly relocated
in Figure 2, the minimum spanning tree is
barely altered.
While Table 4 indicates that the distance
matrices are similar for the deformed and
nondeformed groups, it is also necessary to
consider whether deformation has a significant effect on the multivariate configuration
of nonmetric trait frequencies. To answer
this question, we calculated the 8 x 8 distance matrix between all pairs of deformed
and nondeformed samples. This matrix is
calculated on thirteen traits (asterionic
bone, parietal notch bone, lambdoid os.,
masto-occipital os., infraorbital suture, supraorbital foramen, accessory infraorbital
foramen, divided hypoglossal canal, postcondylar canal, f. ovale open, mastoid f. exsutural, obelionic f., accessory lesser palatine f.) and is shown in Table 5. The
principal coordinates solution in three dimensions (accounting for 92.1% of the distance information) is shown in Figure 3. Inspection of this figure again indicates the
repositioning of the Hopi sample after deformation. Table 6 contains approximate
F-tests for the deformed versus nondeformed crania within each of the four populations. The F-test for these comparisons is:
where N, and N, are the sample sizes within
the deformed and nondeformed samples, “t”
is the number of traits (equal 13 traits in
CRANIAL DEFORMATION AND NONMETRIC TRAITS
1
Nootka
43
Nootk
Fig. 2. Principal coordinates plots with minimum spanning trees for nondeformed (left panel) a n d
deformed (right panel) cranial series.
TABLE 5 . Biological distances between undeformed and deformed (D) series
Hopi
Nootka
Ancon
Kwakiutl
Hopi (D)
Nootka (D)
Ancon (D)
Nootka
Ancon
Kwakiutl
Hopi (D)
Nootka (D)
Ancon (D)
Kwakiutl (D)
0.6374
0.6126
0.2369
1.5348
0.5925
0.7095
0.1162
0.5396
0.4146
1.4003
0.6484
0.1845
0.1911
0.4983
0.5643
0.4907
0.2570
0.0412
0.8366
0.2943
0.2400
1.7071
0.6807
0.8139
0.0744
1.5700
0.5735
0.9430
this case), and the degrees of freedom are “t”
and N, + N, - t - 1. These are approximate tests because there are some missing
data. Rather than using the more exact
methods for missing data analysis (see e.g.,
Morrison, 1990) we have simply substituted
the mean sample sizes rounded to the nearest integer across the thirteen traits. From
Table 6, it is apparent that deformation has
a significant effect on nonmetric cranial
trait frequencies only within the Hopi sample.
DISCUSSION
Our results suggest that while cranial deformation can influence the relative frequency of a minority of nonmetric cranial
traits, the effect is minimal. Certainly, the
overall effect on population distances (and
consequently on any interpretations concerning population structure) is negligible
08
I
‘
not deformed
deformed
04
*
TN
io
Fig. 3. Principal coordinates plot of relationships between eight samples. “ A is Ancon, “ H is Hopi, “ N is
Nootka, a n d “ K is Kwakiutl.
44
L.W. KONIGSBERG ET AL
TABLE 6. Approximate F-tests between deformed and
undefonned samples within populations
Population
Hopi
Nootka
Ancon
Kwakiutl
Number
n.d.
Number s.d.,
d., or m.d.
F-value
P-value
34
21
47
34
60
23
68
75
2.1929
1.4466
1.0234
1.5453
0.0170
0.1960
0.4352
0.1156
for the traits used here. Therefore, it is unlikely that deformation would strongly modify the true population relationships as assessed from cranial nonmetric traits. This
finding is encouraging for the future use of
nonmetric trait variation in the assessment
of population relationships among groups
practicing artificial cranial deformation.
In previous studies, attempts to reconstruct population relationships have been
hampered to a greater or lesser extent by the
presence of cranial deformation. It is occasionally argued that facial and cranial base
measurements can be used to the exclusion
of vault measurements (Corruccini, 1972;
Droessler, 1981; Heathcote, 1986: Table 4.4,
4.5 and his “B25” trait list described on page
157) because the former measurements are
presumably less affected by deformation.
However, a recent study by Cheverud et al.
(1992)based on part of the sample used here
clearly demonstrates that there are both local and global effects of deformation on craniometrics. They find that a number of morphological aspects of the face and cranial
base are affected by deformation (though as
they note in their discussion these effects
are limited and thus could be avoided in biological distance analyses). In contrast, the
results from this study of effects of deformation on nonmetric traits suggest that there
are only local effects on vault ossicles and a
few basicranial foramina, and that these effects are too minor to seriously bias the interpretation of biological distances.
While deformation does not substantially
alter the between-group relationships based
on nonmetric traits, it is not the case that all
cranial nonmetrics are unaffected by all
forms of deformation. Clearly, there are a
few nonmetric traits which are strongly influenced by various forms of deformation. In
particular, coronal ossicles are increased in
relative frequency by all three forms of deformation studied here (fronto-occipital,
lambdoid flattening, and annular). Additionally, there appear to be some localized
effects that are restricted to particular
forms of deformation. For example, the relative frequency of open foramina spinosa and
of foramina of Huschke is increased by
lambdoid flattening (as seen in the Hopi),
though this effect is only present on the
right side. Similarly, there are some very
localized changes in the Kwakiutl cranium
as a result of annular deformation. With this
form of deformation, the relative frequency
of epipteric bones decreases on the left side,
while the relative frequency of parietal
notch bones and masto-occipital ossicles increases on the right side. These three ossicle
locations are all under or near the position of
the wrapping device used to deform the cranium. Given that the majority of deformation effects we have observed relate t o increased frequencies of vault ossicles, these
kinds of traits presumably could be eliminated from study if they were felt to be too
greatly influenced by deformation.
Although it is difficult to comment on trait
etiology from a “naturalistic experiment”
such as this (see, in contrast, Pucciarelli,
1974), we can make some brief observations
on possible developmentaVgeneticpathways
for nonmetrics. In general, our results support the now traditional model (Saunders,
1989) in which discrete traits are partially
heritable characters. We assume that the
traits must not be completely heritable, or
there would be no environmental effect on
their development. The fact that a few traits
are increased o r decreased in frequency under the influence of cranial deformation indicates that heritability is not complete. Beyond this, in the absence of mixed model
complex segregation analyses (Morton and
MacLean, 1974) for discrete cranial traits,
we obviously cannot comment on the relative contribution of polygenic, major gene,
and environmental effects. Nichol (1989)
has recently reported results from complex
segregation analyses of discrete dental
traits, but his analysis did not include the
mixed major gene and polygenic model (Lalouel et al., 1983). Consequently, it is impossible to assess from his study the relative
CRANIAL DEFORMATION AND NONMETRIC TRAITS
contributions of variance components. Such
conclusions must await the study of pedigreed skeletal material using the full mixed
model.
While the ultimate genetic and environmental etiologies cannot be directly specified for the traits studied here, our results
do indicate that deformation can have a
moderate proximate effect on nonmetric
traits. This proximate effect must operate by
redirecting postnatal growth gradients in
the neural capsular matrix, thus altering
bone formation in fontanelles and at sutures, and repositioning morphological elements within skeletal units (see Moss and
Young, 1960; summary in Moore, 1981).
The ability of deformation to produce nonmetric trait alterations must be based on
two factors. First, the outcome of any developmental event which has already occurred
cannot be altered by the later influence of
deformation. As a consequence, traits which
are established prenatally cannot be altered
by postnatal deformation. This is certainly
the case for the traits studied here. All of the
traits whose development is known to occur
specifically in the fetal period (accessory infraorbital foramen, divided hypoglossal canal, postcondylar canal, and the accessory
lesser palatine foramen; see Hauser and De
Stefano, 1989 for developmental timing) are
unaffected by deformation. Second, deformation effects on nonmetric trait frequencies are apparently conditioned on proximity of the trait locations t o the altered
growth gradients. Thus, a number of the
traits whose locations are near areas of maximal growth alteration (e.g., the masto-occipital ossicles, parietal notch bones, epipteric bones, coronal ossicles, and some
basicranial traits) are affected in frequency
when deformation has been applied. That no
facial traits are affected by deformation may
be due either to the fact that most of these
traits are established prenatally, or that
they are located away from the deforming
apparatus.
Our results thus indicate that cranial deformation can affect the relative frequency
of some nonmetric traits, but that this influence is local to the deformation, and that it
only occurs under particular forms of deformation. This finding has two important im-
45
plications. First, the results support the
growing consensus that nonmetric skeletal
traits are the discontinuous morphological
expression of an underlying continuously
distributed trait liability (Corruccini, 1976;
Cheverud et al., 1979; summarized in Saunders, 1989). While this liability is more often
than not heritable (Cheverud and Buikstra,
1981; Sjovold, 19841, it is also true that the
heritabilities are generally much less than
1.O. Consequently, environmental factors
can contribute to the observed phenotypic
variance. In the particular case of artificial
cranial deformation, our results suggest
that some trait liabilities can be influenced
by the local effect of deformation. The second
implication of our results is that nonmetric
trait variation can be used t o recover population relationships even in the face of fairly
extreme artificial cranial deformation.
While deformation does affect the relative
frequency of some nonmetric traits, deformation’s scope of influence on nonmetric
trait variation is much less than that observable for craniometrics. Further, if deformation status can be reliably scored, then
probit analyses such as we have presented
here could be used to correct for the effects of
deformation so that population comparisons
are unaffected by this form of environmental
“noise.”
ACKNOWLEDGMENTS
We thank Dr. Glen Cole and the Field Museum of Natural History and Ms. Jaymie L.
Brauer and the American Museum of Natural History for access to skeletal material.
We also thank Drs. John Blangero, Susan R.
Frankenberg, Sarah Williams-Blangero,
and the anonymous reviewers for their helpful comments on earlier drafts. This research was supported in part by NSF grant
BNS 89-10998 to James M. Cheverud.
LITERATURE CITED
Agresti A (1988) A model for agreement between ratings
on an ordinal scale. Biometrics 44r539-548.
Agresti A, and Agresti BF (1979)Statistical Methods for
the Social Sciences. San Francisco: Dellen Publishing.
Anton SC (1989) Intentional cranial vault deformation
and induced changes of the cranial base and face. Am.
J. Phys. Anthropol. 79:253-267.
Ashford JR, and Sowden RR (1970) Multi-variate probit
analysis. Biometrics 26:535-546.
46
L.W. KONIGSBERG ET AL.
Baughman AL (1988) A FORTRAN function for the bivariate normal integral. Comput. Meth. Prog.
Biomed. 27t169-174.
Bennett KA (1965)The etiology and genetics of wormian
bones. Am. J. Phys. Anthropol. 23~255-260.
Bjork A, and Bjork L (1964) Artificial deformation and
cranio-facial asymmetry in ancient Peruvians. J.
Dent. Res. 43t353-362.
Blackwood B, and Danby PM (1955)A study of artificial
cranial deformation in New Britain. J. Roy. Anthropol. Inst. 85:173-192.
Blangero J , and Williams-Blangero S (1991) Estimating
biological distances from dichotomous threshold traits.
Am. J. Phys. Anthropol. Suppl. 1 2 5 - 5 2 (Abstract).
Boas F (1921) Ethnology of the Kwakiutl: Based on data
collected by George Hunt. 3Ph Ann. Rep. Bur. Amer.
Ethnol. Washington, D.C.: Bur. Amer. Ethnol.
Buikstra J E (1976) Hopewell in the Lower Illinois Valley: A regional study of human biological variability
and prehistoric mortuary behavior. Northwestern
University Archeological Program, Sci. Papers, no. 2.
Cheverud JM,and Buikstra J E (1981) Quantitative genetics of skeletal nonmetric traits in the rhesus
macaques on Cay0 Santiago. Am. J. Phys. Anthropol.
54t43-49.
Cheverud JM, Buikstra JE, and Twichell E (1979) Relationships between non-metric skeletal traits and cranial size and shape. Am. J . Phys. Anthropol. 50t191198.
Cheverud JM, Kohn LAP, Konigsberg LW, and Leigh S
(1992) The effects of fronto-occipital artificial cranial
modification on the cranial base and face. Am. J .
Phys. Anthropol. 88:323-345.
Cole D (1985) Captured Heritage: The Scramble for
Northwest Coast Artifacts. Seattle: University of
Washington.
Corruccini RS (1972) The biological relationships of
some prehistoric and historic Pueblo populations. Am.
J . Phys. Anthropol. 37t373-388.
Corruccini RS (1976) The interaction between nonmetric and metric cranial variation. Am. J . Phys. Anthropol. 44t285-294.
Cybulski JE (1975) Skeletal variability in British Columbia coastal populations: A descriptive and comparative assessment of cranial morphology. Natl.
Mus. Man, Mercury Series. Archaeol. Survey Canada,
Papers, no. 30.
Dennis W, and Dennis MG (1940) Cradles and cradling
practices of the Pueblo Indians. Am. Anthropol.
42:107-115.
Dorsey GA (1895) Crania from the necropolis of Ancon,
Peru. Proc. h e r . Adv. Sci. 43r358-370.
Dorsey GA (1897) Wormian bones in artificially deformed Kwakiutl crania. Am. Anthropol. lOt169-173.
Dorsey GA, and Voth HR (1902) The Mishongnovi Ceremonies of the Snake and Antelope Fraternities. Field
Columbian Museum Anthropol. Ser. vol. 111, no. 3.
Chicago: Field Museum.
Droessler J (1981) Craniometry and Biological Distance:
Biocultural Continuity and Change at the Late-Woodland-Mississippian Interface. Center Amer. Archeol.,
Res. Ser. 1. Evanston: Center Amer. Archeol.
El-Najjar, MY (1978) Maize, malaria and the anemias in
the Pre-Columbian New World. Yrbk. Phys. Anthropol. 20t329-337.
El-Najjar MY, and Dawson GL (1977)The effect of artificial cranial deformation on the incidence of wormian
bones in the lambdoidal suture. Am. J . Phys. Anthropol. 46t155-160.
Gottlieb K (1978)Artificial cranial deformation and the
increased complexity of the lambdoid suture. Am. J.
Phys. Anthropol. 48r213-214.
Green RF, Suchey J, and Gokhale DV (1979) The statistical treatment of correlated bilateral traits in the
analysis of cranial material. Am. J. Phys. Anthropol.
50t629-634.
Hauser G, and De Stefan0 GF (1989) Epigenetic Variants of the Human Skull. Stuttgart: E. Scheizerbart’sche Verlagsbuchhandlung.
Heathcote GM (1986) Exploratory Human Craniometry
of Recent Eskaleutian Regional Groups from the
Western Arctic and Subarctic of North America. BAR
International Series No. 301. Oxford: BAR.
Hellmuth H (1970) Uber den Bau des menschlichen
Schadels bei kunstlicher Deformation. Zeit. Morphol.
Anthropol. 62t30-49.
Hill ID (1973) The normal integral. Applied Statistics
22t424-427.
Kelley TL (1924) Statistical Method. New York: MacMillan Co.
Kohn LAP, and Cheverud JM (1991) The effects of cradleboarding on the cranial base and face. Am. J . Phys.
Anthropol. Suppl. 1 2 ~ 1 0 (Abstract).
6
Konigsberg, LW (19871 Population genetic models for
interpreting prehistoric intra-cemetery biological
variation. Unpub. Ph.D. Diss., Northwestern University, Evanston, 11.
Konigsberg LW, Kohn LAP, Leigh SR, and Cheverud JM
(1991) Cranial modification, nonmetric traits, and biological distance. Am. J. Phys. Anthropol. Suppl.
12t107 (Abstract).
Korey KC (1980) The incidence of bilateral nometric
skeletal traits: A reanalysis of sampling procedures.
Am. J. Phys. Anthropol. 53t19-23.
Lalouel J (1979) GEMINI: A computer program for optimization of a nonlinear function. Dept. Med. Biophys.
Comput., University of Utah, Tech. Rep., no. 14.
Lalouel JM, Rao DC, Morton NE, and Elston RC (1983)
A unified model for complex segregation analysis. Am.
J. Hum. Genet. 35t816-826.
Martin PS, and Willis ES (1940) Anasazi Painted Pottery in Field Museum of Natural History. Anthropology Memoirs vol. 5. Chicago: Field Museum.
McNeill RW, and Newton GN (1965) Cranial base morphology in association with cranial vault deformation.
Am. J . Phys. Anthropol. 23.241-254.
Menzel D (1977) The Archaeology of Ancient Peru and
the Work of Max Uhle. Berkley: RH Lowie Mus. Anthropol.
Moore WJ (1981) The Mammalian Skull. New York:
Cambridge University.
Morrison DF (1990) Multivariate Statistical Methods,
3rd ed. New York: McGraw-Hill.
Morton NE, and MacLean CJ (1974) Analysis of family
CRANIAL DEFORMATION AND NONMETRIC TRAITS
resemblance 111. complex segregation analysis of
quantitative traits. Am. J. Hum. Genet. 26t489-503.
Moss ML (1958) The pathogenesis of artificial cranial
deformation. Am. J. Phys. Anthropol. 16~269-286.
Moss ML, and Young RW (1960)A functional approach
to craniology. Am. J. Phys. Anthropol. 18t281-292.
Nichol CR (1989) Complex segregation analysis of dental morphological variants. Am. J. Phys. Anthropol.
78:37-59.
Olsson U (1979) Maximum likelihood estimation of the
polychoric correlation coefficient. Psychometrika
44~443-460.
Ossenberg NS (1970) The influence of artificial cranial
deformation on discontinuous morphological traits.
Am. J. Phys. Anthropol. 33:357-372.
Pucciarelli HM (1974) The influence of experimental
deformation on neurocranial wormian bones in rats.
Am. J. Phys. Anthropol. 41.29-38.
Reichlen P (1982) Contribucion a1 estudios de las antiguas tecnicas de deformacion cefalica en la costa del
Peru. Revista del Museo Nacional (Lima) 46t381-403.
47
Richstmeier JT, Cheverud JM, and Buikstra J E (1984)
The relationship between cranial metric and nonmetric traits in the rhesus macaques from Cay0 Santiago.
Am. J. Phys. Anthropol. 64~213-222.
Saunders SR (1989) Nonmetric skeletal variation. In
MY Iscan and KAR Kennedy (eds.): Reconstruction of
Life from the Skeleton. New York: Alan R. Liss, pp.
95-108.
Shiprnan JH, Craig E, and Johnston M (1990) Factors
influencing discrete cranial variation among prehistoric Indian groups from San Pedro de Atacama,
Chile. Am. J. Phys. Anthropol. 81t294 (Abstract).
Sjavold T (1984) A report on the heritability of some
cranial measurements and non-metric traits. In GN
Van Vark and WW Howells (eds.): Multivariate Statistical Methods in Physical Anthropology. Boston: D.
Reidl, pp. 223-246.
Sullivan LR (1922) The frequency and distribution of
some anatomical variations in American crania. Anthropol. Papers Am. Mus. Nat. Hist. 23:203-258.
Nootka
not d e f .
deformed.
0.0000 ( 1 3 ) 0.2857 ( 1 4 )
0.0909 i i i j 0.3077 ( 1 3 )
0.0476 ( 2 1 ) 0.2174 ( 2 3 )
0.0455 ( 2 2 ) 0.1818 ( 2 2 )
0.1200 ( 2 5 ) 0.1600 ( 2 5 )
0.1600 ( 2 5 ) 0.1600 ( 2 5 )
0.3529 ( 1 7 ) 0.4762 ( 2 1 )
0.4375 ( 1 6 ) 0.4000 ( 2 0 )
0.0588 ( 1 7 ) 0.1000 ( 2 0 )
0.1111 ( 1 8 ) 0.1111 ( 1 8 )
0.0769 ( 1 3 ) 0.2857 ( 1 4 )
0.0000 ( 1 1 ) 0.3846 ( 1 3 )
0.4348 ( 2 3 ) 0.3333 ( 2 4 )
0.1667 ( 1 2 ) 0.4000 ( 1 5 )
0.4783 ( 2 3 ) 0.2917 ( 2 4 )
0.4167 ( 2 4 ) 0.2800 ( 2 5 )
0.5417 ( 2 4 ) 0.4800 ( 2 5 )
0.6250 ( 2 4 ) 0.6400 ( 2 5 )
0.2400 ( 2 5 ) 0.2400 ( 2 5 )
0.2083 ( 2 4 ) 0.1600 ( 2 5 )
0.4000 ( 2 5 ) 0.3200 ( 2 5 )
0.3200 ( 2 5 ) 0.5200 ( 2 5 )
0.7600 ( 2 5 ) 0.6250 ( 2 4 )
0.7600 ( 2 5 ) 0.4800 ( 2 5 )
0.0400 ( 2 5 ) 0.0400 ( 2 5 )
0.0800 ( 2 5 ) 0.0400 ( 2 5 )
0.0800 ( 2 5 ) 0.0000 ( 2 5 )
0.1200 ( 2 5 ) 0.0000 ( 2 5 )
0.2800 ( 2 5 ) 0.2400 ( 2 5 )
0.3600 ( 2 5 ) 0.2800 ( 2 5 )
0.4211 ( 1 9 ) 0.6087 ( 2 3 )
0.1765 ( 1 7 ) 0.4545 ( 2 2 )
0.8000 ( 2 5 ) 0.7600 ( 2 5 )
0.5600 ( 2 5 ) 0.7200 ( 2 5 )
0.2800 ( 2 5 ) 0.1200 ( 2 5 )
0.3750 ( 2 4 ) 0.2917 ( 2 4 )
0.0417 ( 2 4 ) 0.0000 ( 2 5 )
0.0000 (181 0.0000 ( 1 5 )
0.0588 i 1 7 j o.oooo i i 4 j
Ancon
deformed.
not def.
0.0889 ( 4 5 ) 0.0943 ( 5 3 )
0.0909 ( 4 4 ) 0.0877 (57)
0.1429 ( 4 9 ) 0.2000 ( 7 0 )
0.1250 ( 4 8 ) 0.0845 ( 7 1 )
0.1020 ( 4 9 ) 0.1449 ( 6 9 )
0.1633 ( 4 9 ) 0.1143 ( 7 0 )
0.5000 ( 4 6 ) 0.5797 ( 6 9 )
0.5652 (461 0.5970 ( 6 7 )
0.0426 ( 4 7 ) 0.0149 ( 6 7 )
0.0455 ( 4 4 ) 0.1111 ( 6 3 )
0.0000 ( 3 9 ) 0.0192 ( 5 2 )
0.0000 ( 3 7 ) 0.0000 (50)
0.2917 ( 4 8 ) 0.2571 ( 7 0 )
0.0000 ( 3 7 ) 0.0000 ( 5 0 )
0.2917 ( 4 8 ) 0.2571 ( 7 0 )
0.2708 ( 4 8 ) 0.1806 ( 7 2 )
0.4694 ( 4 9 ) 0.3662 ( 7 1 )
0.3958 ( 4 8 ) 0.2917 ( 7 2 )
0.2708 ( 4 8 ) 0.2535 ( 7 1 )
0.2708 ( 4 8 ) 0.2917 ( 7 2 )
0.2128 ( 4 7 ) 0.2429 ( 7 0 )
0.3404 ( 4 7 ) 0.2676 ( 7 1 )
0.8980 ( 4 9 ) 0.8889 ( 7 2 )
0.8571 ( 4 9 ) 0.8219 ( 7 3 )
0.0213 ( 4 7 ) 0.0143 ( 7 0 )
0.0208 ( 4 8 ) 0.0000 ( 7 0 )
0.0417 ( 4 8 ) 0.0986 (71)
0.1458 ( 4 8 ) 0.0714 ( 7 0 )
0.4082 ( 4 9 ) 0.3973 ( 7 3 )
0.3878 ( 4 9 ) 0.3662 (71)
0.4792 ( 4 8 ) 0.4286 ( 7 0 )
0.3878 ( 4 9 ) 0.4429 ( 7 0 )
0.6531 ( 4 9 ) 0.6164 ( 7 3 )
0.5510 ( 4 9 ) 0.6301 ( 7 3 )
0.3043 ( 4 6 ) 0.4179 ( 6 7 )
0.2609 ( 4 6 ) 0.4545 (66)
0.0000 ( 4 9 ) 0.0274 ( 7 3 )
0.0000 ( 4 1 ) 0.0339 ( 5 9 )
0.1628 ( 4 3 ) 0.3443 ( 6 1 )
Kuakiut 1
not def.
deformed.
0.2941 ( 1 7 ) 0.0882 ( 3 4 )
0.1176 i i 7 j 0.0556 ( 3 6 )
0.1714 ( 3 5 ) 0.1538 ( 7 8 )
0.1875 ( 3 2 ) 0.1667 ( 7 8 )
0.2703 ( 3 7 ) 0.2469 ( 8 1 )
0.1892 ( 3 7 ) 0.2593 ( 8 1 )
0.6000 ( 2 5 ) 0.6618 (68)
0.5600 ( 2 5 ) 0.5857 ( 7 0 )
0.0714 ( 2 8 ) 0.2031 (64)
0.0000 ( 2 6 ) 0.1493 ( 6 7 )
0.1053 ( 1 9 ) 0.3409 ( 4 4 )
0.0556 ( 1 8 ) 0.4130 ( 4 6 )
0.3429 ( 3 5 ) 0.4815 ( 8 1 )
0.3056 ( 3 6 ) 0.3846 ( 7 8 )
0.5405 ( 3 7 ) 0.5122 ( 8 2 )
0.6216 ( 3 7 ) 0.4634 ( 8 2 )
0.3333 ( 3 6 ) 0.2927 ( 8 2 )
0.3889 ( 3 6 ) 0.2840 ( 8 1 )
0.1389 ( 3 6 ) 0.1585 ( 8 2 )
0.2222 ( 3 6 ) 0.2152 ( 7 9 )
0.8611 ( 3 6 ) 0.8642 ( 8 1 )
0.7778 ( 3 6 ) 0.7625 ( 8 0 )
0.0278 ( 3 6 ) 0.0366 ( 8 2 )
0.0278 ( 3 6 ) 0.0610 ( 8 2 )
0.0278 ( 3 6 ) 0.0610 ( 8 2 )
0.1111 ( 3 6 ) 0.0482 ( 8 3 )
0.2222 ( 3 6 ) 0.1687 (831
0.3056 ( 3 6 ) 0.1446 ( 8 3 )
0.4516 ( 3 1 ) 0.6250 ( 7 2 )
0.4000 ( 3 0 ) 0.5600 ( 7 5 )
0.5676 ( 3 7 ) 0.6429 (84)
0.4865 ( 3 7 ) 0.6190 (84)
0.5429 ( 3 5 ) 0.5696 (791
0.4444 ( 3 6 ) 0.5500 ( 8 0 )
0.0000 ( 3 7 ) 0.0238 ( 8 4 )
0.0000 ( 2 7 ) 0.0000 ( 6 0 )
0.1538 ( 2 6 ) 0.1846 ( 6 5 )
0.0556 (361 0.0482 (831
0.0526 ( 1 9 ) 0.0392 ( 5 1 1
Appendix Fig. A. Nonmetric trait frequencies in nondeformed crania and deformed crania (left sides
listed before right sides). Sample sizes (observable number of crania per trait) are shown in parentheses.
Hopi
deformed.
not def.
0.1429 ( 2 8 ) 0.0588 ( 5 1 )
0.0400 i 2 5 j 0.1224 ( 4 9 )
0.3235 ( 3 4 ) 0.3881 ( 6 7 )
Aster ioni c bone
0.2286 ( 3 5 ) 0.2576 (66)
0.0513 ( 3 9 ) 0.0294 (68)
P a r i e t a l notch bone
0.0789 ( 3 8 ) 0.0448 ( 6 7 )
0.3333 ( 3 3 ) 0.5333 ( 6 0 )
Lambdoid 0s.
0.3636 ( 3 3 ) 0.4500 ( 6 0 )
0.3030 ( 3 3 ) 0.2000 ( 5 5 )
Masto-occipital 0s.
0.3529 ( 3 4 ) 0.2308 ( 5 2 )
0.0000 ( 2 8 ) 0.0500 ( 6 0 )
Coronal 0s.
0.0000 ( 2 9 ) 0.0545 ( 5 5 )
0.4722 ( 3 6 ) 0.4531 (64)
I n f r a o r b i t a l suture
0.0000 ( 2 9 ) 0.0545 ( 5 5 )
Supraorbital foramen 0.4722 ( 3 6 ) 0.4531 (64)
0.3750 ( 3 2 ) 0.4154 ( 6 5 )
Access. infraorb. f . 0.2162 ( 3 7 ) 0.4348 ( 6 9 )
0.3611 ( 3 6 ) 0.3188 ( 6 9 )
0.0556 ( 3 6 ) 0.1311 ( 6 1 )
Divided hypoglossal
0.1563 ( 3 2 ) 0.0635 ( 6 3 )
0.1389 ( 3 6 ) 0.0746 ( 6 7 )
Postcondylar canal
0.0833 ( 3 6 ) 0.1667 (66)
0.9143 ( 3 5 ) 0.9538 ( 6 5 )
F. ovate open
0.9032 ( 3 1 ) 0.9219 (64)
0.0278 ( 3 6 ) 0.0159 ( 6 3 )
F. spinosum open
0.0000 ( 3 4 ) 0.0441 (68)
0.1143 ( 3 5 ) 0.1231 ( 6 5 )
F. Huschke
0.0294 ( 3 4 ) 0.1029 (68)
Mastoid f . exsutural 0.2857 ( 3 5 ) 0.3231 ( 6 5 )
0.3056 ( 3 6 ) 0.2381 ( 6 3 )
0.1842 ( 3 8 ) 0.1290 ( 6 2 )
Obelionic foramen
0.1389 ( 3 6 ) 0.2031 (64)
Access. less. palat. 0.4615 ( 3 9 ) 0.5362 ( 6 9 )
0.3846 ( 3 9 ) 0.5072 ( 6 9 )
0.2727 ( 3 3 ) 0.3548 ( 6 2 )
Metopic suture
0.2188 ( 3 2 ) 0.4098 ( 6 1 )
Bregmat ic bone
0.0000 ( 3 9 ) 0.0290 ( 6 9 )
Apical bone
Inca bone
0.0000 ( 3 2 ) 0.0000 ( 6 1 )
S a g i t t a l 0s.
0.0625 ( 3 2 ) 0.4000 ( 6 0 )
Trait
E .D i ,Dter i C bone
Документ
Категория
Без категории
Просмотров
3
Размер файла
1 094 Кб
Теги
deformation, cranial, variation, traits, nonmetric
1/--страниц
Пожаловаться на содержимое документа