close

Вход

Забыли?

вход по аккаунту

?

Effects of annular cranial vault modification on the cranial base and face.

код для вставкиСкачать
AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 90:147-168 (1993)
Effects of Annular Cranial Vault Modification on the Cranial Base
and Face
LUCI ANN P. KOHN, STEVEN R. LEIGH, SUSAN C. JACOBS, AND
JAMES M. CHEVERUD
Department of A m t o m y and Neurobiology (L.A.P.K., J.M.C.) and
Division of Biological Sciences (S.C.J.1, Washington University School of
Medicine, St. Louis, Missouri 63110, and Department of Anthropology,
Northwestern University, Euanston, lllinois 60201 (S.R.L.)
KEY WORDS
Cranial growth, Finite element scaling, Kwakiutl,
Nootka, Artificial modification
ABSTRACT
Artificial modification of the cranial vault was practiced by a
number of prehistoric and protohistoric populations, frequently during a n
infant’s first year of life. We test the hypothesis that, in addition to its direct
effects on the cranial vault, annular cranial vault modification has a significant indirect effect on cranial base and facial morphology.
Two skeletal series from the Pacific Northwest Coast, which include both
nonmodified and modified crania, were used: the Kwakiutl (62 nonmodified,
45 modified) and Nootka (28 nonmodified, 20 modified). Three-dimensional
coordinates of 53 landmarks were obtained using a diagraph, and 36 landmarks were used to define nine finite elements in the cranial vault, cranial
base, and face. Finite element scaling was used to compare average nonmodified and average modified crania, and the significance of the results were
evaluated using a bootstrap test.
Annular modification of the cranial vault produces significant effects on the
morphology of the cranial base and face. Annular modification in the Kwakiutl resulted in restrictions of the cranial vault in the medial-lateral and
superior-inferior dimensions and a n increase in anterior-posterior growth.
Similar dimensional changes are observed in the cranial base. The Kwakiutl
face is increased anterior-posteriorly and reduced anterior-laterally to posterior-medially. Similar effects of modification are observed in the Nootka cranial vault and cranial base, though not in the face. These results demonstrate
the developmental interdependence of the cranial vault, cranial base, and
face. 01993Wiley-Liss, Inc.
Artificial cranial modification was practiced by a number of prehistoric and protohistoric North and South American Indian
and Pacific Island populations. This practice
had marked effects on cranial morphology
(Dorsey, 1895, 1897; Hrdlicka, 1905;
Dingwall, 1931; Morton, 1939; Blackwood
and Danby, 1955; Bjork and Bjork, 1964;
McNeill and Newton, 1965; Brothwell, 1975;
Schendel et al., 1980). Typically, cranial
modification was practiced during the first
year of life and the use of modifying appliances was commonly restricted to the cra0 1993 WILEY-LISS, INC
nial vault. Crania from these individuals
provides a test of relationships between the
cranial vault, cranial base, and face during
growth. In this analysis, we test the independence of growth between cranial regions.
Specifically, we evaluate whether or not
modification of the cranial vault has a sig-
Received December 11,1991; accepted J u n e 28,1992.
Address reprint requests to Luci Ann P. Kohn, Department of
Anatomy and Neurobiology, Box 8108,660 S. Euclid Ave., Washington University School of Medicine, St. Louis, MO 63110.
148
L.A.P. KOHN ET AL
nificant effect on the growth of the cranial
base and face.
Intentional artificial modification has
been classified into two major types, anterior-posterior and annual or circumferencial. This classification is primarily based on
external vault morphology (Dingwall, 1931).
Anterior-posterior modification, a s practiced at Ancon (in Peru) or Makapuan (Hawaii), or by the Songish (Pacific Northwest
coast), among others, results from pressure
exerted on the frontal and occipital by either
a cradle board or a headdress (Boas, 1921;
Dingwall, 1931; Schendel et al., 1980; Allison et al., 1981). These appliances restrict
anterior-posterior growth, resulting in crania that appear short in the anterior-posterior dimension and wide in the medial-latera1 dimension (Boas, 1921; Dingwall, 1931;
Anton, 1989; Cheverud and Midkiff, 1992;
Cheverud et al., 1992). Annular or circumferencial cranial modification was practiced
by many groups, including the Kwakiutl and
Nootka (both in the Pacific northwest coast),
the Peruvian Indians (reported in Anton,
1989), and the Arawe (Blackwood and
Danby, 1955). Annular modification was
achieved by circumferentially wrapping the
cranial vault (Dingwall, 19311, a practice
that resulted in a skull with a flattened forehead and a posterior elongation of the superior portion of the cranial vault, including
the frontal, parietals, and occipital regions.
Modified individuals from these groups
were often characterized as having “loafs h a p e d heads (Dingwall, 1931). The effect
of annular modification is illustrated in Figure 1.
Several studies have examined the effects
of cranial vault modification on the morphology of the cranial base and face, concentrating on the effects of cranial vault modification on the cranial base angle (Oetteking,
1924; Moss, 1958; McNeill and Newton,
1965; Anton, 1989). Most studies found that
both anterior-posterior (Oetteking, 1924;
McNeill and Newton, 1965; Anton, 1989)
and annular modification (McNeill and
Newton, 1965; Anton, 1989) resulted in a
greater angle between the anterior and posterior cranial base. These studies show that
the cranial base is affected by cranial vault
modification; however, it is unclear how the
Fig. 1. Annular modification in the Kwakiutl (specimen number 40797, Field Museum of Natural History,
Chicago): a, lateral view, B, superior view.
angular change noted is affected, i.e., which
portion of the cranial base is changed morphologically.
It is also unclear from previous studies
whether there is a relationship between artificial modification and facial morphology.
Some studies report that facial dimensions
are unaffected by artificial modification
(Rothhammer, 19841, but others found that
artificial modification did influence facial
dimensions (Oetteking, 1930; Blackwood
and Danby, 1955; Cybulski, 1975; Schendel
et al., 1980; Anton, 1989; Cheverud et al.,
1992). In general, annular cranial vault
modification seems associated with an increase in facial height and a decrease in facial breadth (Blackwood and Danby, 1955;
Cybulski, 1975; Anton, 1989). The upper
face and orbital regions seem to be the principally influenced regions. Inconsistent effects of annular modification were observed
in the palatal region, with reduced palate
ANNULAR CRANIAL. VAULT MODIFICATION
149
length (Cybulski, 1975; Anton, 1989) and may therefore be insufficient to measure difbreadth (Anton, 1989)observed in some pop- ferences between nonmodified and modified
ulations and not others (Blackwood and crania.
Danby, 1955).
These problems combine to suggest that
The specific effects of cranial vault modifi- the morphological bases of modification are,
cation on cranial base and facial morphology in actuality, rather poorly known. We sugare difficult to discern from the previous gest that much of the uncertainty in interstudies. This problem is related to heteroge- pretation of previous results may be reneous sample composition and choice of solved through new methods of analysis.
measurements (Cheverud et al., 1992). Cheverud et al. (1992) utilize a n alternative
Cheverud et al. (1992) indicate that most of method, the finite element scaling method,
the studies cited above compare nonmodi- to approach the general problem of cranial
fied and modified individuals from different modification. Specifically they analyze difpopulations. For example, Oetteking (1924) ferences between nonmodified and anteriorcompared nonmodified Chumash Indians posterior modified crania in prehistoric crafrom California to anterior-posterior modi- nia from Ancon and the protohistoric
fied Chinook from the Pacific Northwest Songish, finding significant differences becoast. Bjork and Bjork (1964) and Anton tween nonmodified and modified crania
(1989) analyzed nonmodified and modified along dimensions not usually considered in
crania from various different coastal and standard morphometric analyses. They also
highland Peruvian groups. These compari- found different patterns of modification besons confound interpopulation morphologi- tween the two skeletal series and suggested
cal differences with the effects of modifica- that differences between the effects of frtion. Of the previously cited studies, only onto-occipital modification may be due to
Schendel et al. (1980) and Cheverud et al. differences in modifying appliances.
In this report we follow Cheverud et al.
(1992) included nonmodified and modified
individuals that were from the same skele- (1992) and present a n analysis of the effect
of annular cranial reshaping on the mortal series.
Additional uncertainty arises from char- phology of the cranial base and face in the
acters used to compare nonmodified and Kwakiutl and Nootka. We use finite element
modified crania. Most studies either use tra- scaling methods to describe the differences
ditional distances or angles between ana- between nonmodified and modified crania
tomical landmarks on crania (Oetteking, within each sample and a nonparametric
1924; Blackwood and Danby, 1955; Bjork bootstrapping test to evaluate statistical
and Bjork, 1964; McNeill and Newton, 1965; significance. Based on previous studies, we
Cybulski, 1975; Schendel et al., 1980; An- expect to find that annular cranial vault
ton, 1989) or X-rays (Moss, 1958; Anton, modification will have significant effects on
1989). Many traditional lengths (e.g., facial the morphology of the length, breadth and
height, or nasion to gnathion) and angles height of the cranial base and face. We ex(e.g., sella to nasion to gnathion) cover re- pect these regions to exhibit increased
gions of the cranium displaying different length and height, and reduced breadth. We
growth dynamics. Whereas many of the contrast the effects of annular and anteriormeasures used represent dimensional dif- posterior modification on the morphology of
ferences between nonmodified and modified the cranial base and face.
individuals, it is often difficult to discern
which specific cranial locations are modified
MATERIALS AND METHODS
(Moyers and Bookstein, 1979; Cheverud et
Samples
al., 1983; Cheverud and Richtsmeier, 1986;
Richtsmeier and Cheverud, 1986). MoreThe samples analyzed in this study are
over, the traditional measurements may not housed at the Field Museum of Natural Hisbe along the dimensions that best discrimi- tory, Chicago, and the American Museum of
nate the nonmodified and modified individ- Natural History, New York City. The Kwauals (Cheverud and Richtsmeier, 1986) and kiutl and Nootka series are both from the
150
L.A.P. KOHN ET A L
Pacific Northwest coast, and both tribes belong to the Wakashan linguistic subgroup.
They represent distinct groups and probably
did not extensively intermarry (Cybulski,
1975). Both populations practiced annular
cranial modification, and nonmodified and
modified individuals were included within
both museum series. The inclusion of both
nonmodified and modified individuals from
a single series provides a control for interpopulation variation in craniofacial morphology, a source of variation not usually
considered in previous studies (see above).
Comparisons of modified Kwakiutl with
modified Nootka test whether annular cranial modification represents a homogeneous
class of modification across populations.
Within each sample cranial morphology
was scored a s either “not modified (or nonmodified, nm) if there was no evidence of
cranial modification, “slightly modified
(sm) if there was some indication of modification (especially asymmetry), “modified”
(m) if the cranial vault had clearly undergone modification, or “much modified” (mm)
if cranial modification was extreme. Intraand interobserver reliability in scoring for
modification was found to be high in the
Field Museum of Natural History series
(Konigsberg et al., 1992). Cranial modification was scored by one individual at the
American Museum of Natural History. Estimates of observer error in modification scoring were not undertaken for this sample.
Kwakiutl
The Kwakiutl sample includes 59 adult
males (34 not modified, 25 modified; sm, m,
or mm) and 48 adult females (28 not modified, 20 modified; sm, m, or mm). Gender
identification of samples from the Field Museum of Natural History was based on museum records, and cranial rugosity and morphology were used for samples at the
American Museum of Natural History. The
collections include protohistoric and early
historic specimens collected by Franz Boas,
George Dorsey, George Hunt, and C. F.
Newcombe near the turn of the century.
Annular modification among the Kwakiutl was achieved by wrapping a n infant’s
head with kelp. Although cranial modifica-
tion was carried out on both males and females, the kelp binding was kept on girls’
heads for a longer period of time, and the
binding was changed more often in girls
than in boys (Boas, 1921). Regional differences were found in intensity and duration
of modification (Boas, 1921). Some groups
bound the infant’s head for 4 months, but
others practiced binding for as much as a
year. Some groups changed the band every 8
days; others kept it on for 12 days (Dingwall,
1931). The variation in intensity of cranial
modification may reflect geographic differences. Nonmodified crania may represent
either geographic (Dingwall, 1931) or temporal (Boas, 1921) differences in cranial
vault modification practices,
Nootka
The Nootka sample includes 29 adultmales (17 not modified, 12 modified; sm, m,
mm) and 19 adult females (11not modified,
8 modified; sm, m, mm). The Nootka samples were collected by Franz Boas, George
Dorsey, and C. F. Newcombe near the turn
of the century (Cole, 1985) and probably also
date to postcontact times.
The annular cranial modification technique practiced by the Nootka differed from
that practiced by the Kwakiutl. The Nootka
also bound a n infant’s cranial vault with a
fibrous wrap, and the infants were kept on a
cradle with a head presser and pads until
they were able to walk (Dingwall, 1931). Sapir (1922) commented that the Nootka did
not “like big foreheads” (i.e., unmodified
heads). As with the Kwakiutl, unmodified
crania within the samples may represent
temporal differences in cranial vault modification (Dingwall, 1931).
Sample classification
The relationship between cranial form
and our classification system was evaluated
through discriminant function analysis using arcs and chords between cranial vault
landmarks. This procedure provided insight
into the reliability of our scoring system. Results of discriminant function analysis suggested that nonmodified and modified could
be reliably separated (P < 0.05). Five Kwakiutl crania were misclassified according to
ANNULAR CRANIAL VAULT MODIFICATION
the discriminant function analyses and they
had high posthoc probabilities of belonging
to alternate groups. One nonmodified individual was reclassified to modified and four
modified individuals were reclassified to
nonmodified. Discriminant function analysis was also used to determine if the slightly
modified individuals could be reliably classified as either nonmodified or modified. The
slightly modified individuals were not used
to generate the discriminant function, but a
probability was calculated for their unbiased classification. All of the slightly modified Kwakiutl and Nootka individuals could
be reliably classified into the nonmodified
(N~,.,&i~t]
= 26, NNootka= 5 ) Or modified
(NKwakiutl = 8, "ootka
= 4, soups.
Measurements
The three-dimensional coordinates of 53
landmarks (Fig. 2, Table 1) were obtained
from the 107 Kwakiutl and 48 Nootka crania
using a diagraph. All coordinate data were
collected by one individual (S.R.L.). The Xand Y-coordinates were entered into a computer using a Tektronix 2-dimensional digitizer, and the 2 coordinate was entered from
the computer keyboard. After data entry,
the coordinates were registered or oriented
with the origin at the anterior nasal spine.
Three arbitrary axes were defined: (1) an
anterior-posterior axis was defined connecting anterior nasal spine to lambda; (2) a medial-lateral axis, orthogonal to the first axis
and passing roughly through the right and
left external auditory meatus; (3)a superiorinferior axis, approximately parallel to a
bregma-opisthion line, orthogonal to the
first two axes. These axes aid in screening
data for outliers and describing directions of
shape change, and they do not affect finite
element scaling analysis.
Finite element scaling analysis was used
to measure the morphometric differences
between nonmodified and modified crania
within each sample. Finite element scaling
is a morphometric technique that allows interpretation of differences in form. More
specifically it measures the nonhomogeneous deformation of one form, an initial or
reference form into a target or subsequent
form (Bookstein, 1978, 1983, 1984, 1987;
151
Lewis et al., 1980; Cheverud et al., 1983;
Moss et al., 1985; Cheverud and Richtsmeier, 1986; Lozanoff and Diewert, 1986;
Richtsmeier and Cheverud, 1986; Moss,
1988). The finite element scaling analyses
were performed using SCAL3D (Hamner
and Bachrach, 19861,and the accompanying
graphics were generated by FIESCA (Morris, 1989).
This method is based on differences in
three-dimensional geometric representations of forms. Individual crania are subdivided into a number of subsidiary geometric
forms called finite elements. In our application, a finite element is formed by connecting nodes or landmarks by straight lines.
Elements are used t o depict anatomical regions, and elements are defined so as to describe the form being studied. In a threedimensional object, finite elements may be
defined by four (tetrahedral), five (pyramidal), six (wedge-shape)or eight (hexahedral)
landmarks. It can be noted that the finite
element scaling analyses of tetrahedral and
hexahedral finite elements are equivalent.
The form change at a node of a hexahedron
is the same as for the tetrahedron formed by
the node and the other nodes to which it is
connected in a hexahedron (Cheverud et al.,
1992). Thirty-six landmarks were used to
define nine hexahedral finite elements, depicting the upper and lower face and cranial
vault on both the right and left sides, and
the cranial base (Fig. 2, Table 2 ) .
Finite element scaling analysis produces
a form (or Lagrangian) strain tensor for each
landmark. The form strain tensor is a symmetric matrix with rows and columns equal
to the number of dimensions in the analysis.
The magnitudes and directions of deformation are derived from this form strain tensor, as the principle values and principle directions, respectively. The principal values
(ei) provide the magnitudes of the largest,
intermediate, and smallest orthogonal increases necessary to deform the initial form
into the target form. The principal directions (Pi) describe the orientations along
which this form change occurs. The principal directions are described with reference
to the initial form and along the axes previously described.
L.A.P. KOHN ET AL.
152
A
B
24
Y
L
Fig. 2. Location of landmarks and finite elements.
Landmarks listed in Table 1 are identified by number
and the nine finite elements are listed in Table 2. The
nine finite elements, representing the upper and lower
anterior cranial vault, posterior cranial vault, upper and
lower face, and cranial base are represented by lines
connecting the appropriate landmarks. A. Lateral view
of the cranium. B. Inferior view of the cranium. Only
landmarks on the left side are represented in this lateral
view since bilateral landmarks would appear superimposed. The element representing the cranial base includes landmarks from both the right and left sides, and
these landmarks, (16,171,
(30,311,
(43,441,
(45,461,
and
(30,31,5 3 ) , appear superimposed in this figure (Cheverud et al., 1992).
The form change that occurs in deforming
the reference into the target can be decomposed into its component parts of size (in-
crease or decrease) and shape change (Fig.
3). If one considers a sphere of standard size
located at each landmark in the reference
153
ANNULAR CRANIAL VAULT MODIFICATION
TABLE 1. Landmarks recorded for the Kwakiutl and
Nootka crania
TABLE 2. Fznate elements used in the analysis of the
effects of annular on Kwakiutl and Nootka crania
Landmarks
Finite Element
1. Intradentale Superior
2. Premaxilla l(R) anterior alveolar ridge between the
canine and premolar1
3. Premaxilla (L)
4. Posterior Nasal Spine
5. Nasale
6. Zygomaxillare superior (R)
7. Zygomaxillare superior (Lj
8. Zygomaxillare inferior (R)
9. Zygomaxillare inferior (Rj
10. Infratemporal Crest (R)
11. Infratemporal Crest (L)
12. Vomer Spine
13. Nasion
14. Frontomalare (R)
15. Frontomalare (L)
16. Pterion (R)
17. Pterion (L)
18. Optic foramen (R)
19. Optic foramen (L)
20. Bregma
21. Bregma-nasion (point halfway along hregma-nasion
arc)
22. Bregma-Pterion [(R), point halfway along bregmapterion arcl
23. Bregma-Pterion (L)
24. Lambda
25. Asterion (R)
26. Asterion (L)
27. Bregma-lambda (point halfway along bregma-lambda
arcl
28. Bregma-asterion [(R), point halfway along bregmaasterion arcl
29. Bregma-asterion (Ll
30. Pterion-asterion [(R), point halfway along pterionasterion arcl
31. Pterion-asterion (Ll
32. Pterion-lambda [(R), point halfway along pterionlambda arcl
33. Pterion-lambda (L)
34. Lambda-asterion “R), point halfway along
lambda-asterion arcl
35. Lambda-asterion (L)
36. Opisthion
37. Basion
38. Lambda-opisthion (point halfway along lambdaopisthion arcl
39. External auditory meatus (R)
40. External auditory meatus (L)
41. Temporo-sphenoid (R)
42. Temporo-sphenoid (L)
43. Jugular process (R)
44. Jugular process (L)
45. Foramen lacerum (R)
46. Foramen lacerum (L)
47. Anterior Nasal Spine
48. Maxillary tuberosity (R)
49. Maxillary tuberosity (L)
50. Zygomatic arch (Rl
51. Zygomatic arch (L)
52. Optic foramen midpoint (average of 18 and 19)
53. Pterion-asterion midpoint (average of 30 and 31)
Lower face (R)
Lower face (L)
Upper face (R)
Upper face (L)
Cranial base
Posterior vault (R)
Posterior vault (Ll
Anterior vault (R)
Anterior vault (L)
form, size change measures the degree to
which the sphere increases or decreases in
volume during the deformation of the refer-
Landmarks
1, 2, 4, 5, 6 , 10, 12, 48
1, 3, 4, 5, 7, 11, 12, 49
5, 6 , 10, 12, 13, 14, 16, 52
5, 7, 11, 12, 13, 15, 17, 52
16, 17, 30, 31, 43, 44, 45, 46
24, 25, 28, 30, 36, 38, 43, 53
24, 26, 29, 31, 36, 38, 44, 53
16, 20, 22, 24, 25, 27, 28, 53
17, 20, 23, 24, 26, 27, 29, 53
Landmarks are defined in ‘Fable 1
ence into the target form. Size change, s, in
the region surrounding each landmark is estimated as the average of the principle values once they are transformed to linear scale
by the equation,
That is, size change is
s =
(L,
+ L, + L,)/3
Shape change measures the degree to which
the standard sphere is deformed into an ellipsoid during the deformation of the reference into the target. Shape change, t, in the
region surrounding each landmark is estimated as the standard deviation of the principal values transformed to linear scale.
That is,
t
=
{[(L, - Sl2 + (L, - Sl2 + (L3 - s)”/31}1”
Shape change will be large if size increases
substantially along different anatomical
axes. Conversely, shape change will be minimal when size changes are approximately
equal in all directions. Global size change is
defined here as the local size change averaged over all landmarks. Global shape
change is the standard deviation of local size
change measured at each landmark (Cheverud and Richtsmeier, 1986; Richtsmeier
and Cheverud, 1986). The proportional
change in length along the particular anatomical dimensions specified is measured by
the antilog of Li.
Statistical tests
Multivariate analysis of variance of arcs
and chords between cranial vault land-
154
L.A.P. KOHN ET AL.
FORM
=
ASIZE
+ ASHAPE
P1
A SIZE ONLY
P1
P1
ASHAPE ONLY
PI
Fig. 3. The form strain tensor produced by finite element scaling is represented by an ellipse. For each landmark in the comparison of the reference and target
form, this ellipse represents the local change in size and
shape in the region surrounding each landmark. The
tensor represents the degree to which a circle located at
each landmark is deformed into an ellipse in the transformation. P1 and P2 represent the principal axes of
deformation in this two-dimensional presentation. The
first figure illustrates form change as the sum of size
and shape change. The second figure illustrates a transformation in which only size increase occurs, and all
anatomical dimensions increase equally. The third figure represents a transformation with only shape
change. Whereas there is no total change in area, some
anatomical dimensions increased in the transformation
and other dimensions are decreased (Cheverud et al.,
1992).
marks was used to test whether cranial
vault modification showed significant effects of sex, modification, and a n interaction
of sex and modification. Sex and modification showed no significant interaction, suggesting that males and females did not react
differently to cranial modification. I n the
absence of a significant interaction, data
from males and females can be pooled to test
the effect of modification on the cranial
vault, cranial base, and face.
We test the hypothesis that within each
series there is no difference between the average nonmodified and the average modified individual. In other words, we test
whether or not the modified sample could
represent a sample drawn randomly from
the nonmodified sample. Observed differences between nonmodified and modified
averages are measured through finite element scaling. A weighted average of the
nonmodified individuals was calculated
such that the sex ratio of the nonmodified
group was equal to that in the modified
group. Within each series, the average nonmodified individual was deformed into the
average modified individual using finite element scaling. The significance of this difference was tested by a nonparametric bootstrap procedure suggested by Lele and
Richtsmeier (1991) and used by Cheverud et
al. (1992). By this procedure, bootstrap nonmodified and modified averages are obtained by random sampling with replacement from the nonmodified sample. Within
the Kwakiutl and Nootka, the random sampling is stratified by sex, such that males
and females are drawn from the nonmodified sample in the sex ratio observed in the
modified sample. The bootstrap nonmodified average is deformed into the bootstrap
modified average. This bootstrap procedure
is repeated a large number of times. These
results are based on 500 bootstrap samples.
The statistical significance of the difference
between nonmodified and modified crania is
measured by the proportion of bootstrap
samples from the total of 500 for which differences exceed those observed between the
groups. This procedure is used to test the
significance of local size and shape change,
global size and shape change, and element
volume differences. The tests of local and
global size change, and element volume differences are two-sided tests, and probabilities <0.025 or >0.975 denote significance a t
the 5% level. The tests of local and global
shape change are one-sided tests, and probabilities less than 0.05 are significant. It
should be noted that the directions of local
shape change should not be interpreted unless there is a significant local shape difference (Bookstein, 1984; Cheverud et al., 1992).
Local shape change is significant only when
size change differs along anatomical axes.
ANNULAR CRANIAL VAULT MODIFICATION
155
B
Fig. 4. Superimposition of average normal (solid line) and modified Kwakiutl (dotted line) following
resizing and Procrustes rotation of all landmarks: A, lateral view, B, inferior view.
RESULTS
Kwakiutl
Lateral and basal views of the average
nonmodified and modified Kwakiutl are superimposed after generalized procrustes rotation (Goodall and Bose, 1987) in Figure 4.
This illustrates the differences between the
average nonmodified and modified Kwa-
kiutl and aids in interpretation of the finite
element scaling results. Generalized procrustes rotation minimizes the sum of the
squared differences between homologous
landmarks in the two forms once the average nonmodified and modified Kwakiutl are
superimposed.
The average modified male Kwakiutl was
12% smaller in total volume than the aver-
156
L.A.P. KOHN ET AL
age nonmodified male Kwakiutl, and the total volume of the average modified female
was 1.6%larger than the average nonmodified female Kwakiutl. This means that,
prior to analysis, the crania should be rescaled to ensure that differences between
nonmodified and modified Kwakiutl are not
due solely to their difference in volume. All
coordinates of the modified crania were resized prior to finite element scaling analysis. The results discussed below therefore
represent analyses after crania were scaled.
As required by our methods, there was no
global size difference between the nonmodified and modified Kwakiutl.
The finite element scaling results measuring the differences between nonmodified
and modified Kwakiutl are presented in Table 3 and Figure 5. Whereas the face, cranial
base, and posterior cranial vault are smaller
in the average modified cranium, only the
smaller volume of the upper face approaches
significance. Global shape differences approach significance, indicating variation
among regions in deviation from nonmodified.
A significant effect of the deforming device on the cranial vault is present. When
compared to the average nonmodified
Kwakiutl, the average modified Kwakiutl
exhibits a significant local increase in size
along the midline, at bregma-lambda,
lambda, and the pterion-asterion midpoint.
There is a significant decrease in size at the
left bregma-pterion, right and left bregmaasterion, right pterion-asterion, and left asterion. There is a significant shape change
at all landmarks in the cranial vault with
the exception of right and left pterion-asterion. Right and left pterion, and the landmarks located along the cranial midline, exhibit an increase of about 8% along the
anterior-posterior axis, while decreasing 2%
medial-lateral and 5% superior-inferiorly.
Most of the remaining lateral landmarks increase about 6% along an axis from anteriorlateral to posterior-medial and decrease 5%
along superior-inferior and anterior-medial
to posterior-lateral directions. These results
suggest that annular modification of the
cranial vault causes the cranial vault t o
grow longer (anterior-posterior axis), narrower (medial-lateral axis), and shallower
(superior-inferior axis) compared to nonmodified individuals from the same sample.
Annular cranial vault modification affects
the cranial base, although the effect is less
than was seen on the cranial vault. The average modified Kwakiutl is significantly
smaller at the midpoint between the right
and left optic foramen. There is a significant
shape change at the right and left jugular
processes and at the vomer spine. The jugular processes show a 7.5% anterior-posterior
increase and 3% decreases medial-laterally
and superior-inferiorly. The vomer spine increases 3% along the anterior-posterior axis
and decreases 3% in the medial-lateral direction and 2% in a superior-inferior direction. Nonsignificant changes were similar to
the significant ones. These findings imply
that annular modification of the cranial
vault causes anterior-posterior increase of
the cranial base and reduction along mediallateral and superior-inferior dimensions.
Annular cranial vault modification appears to have small but significant effects on
the Kwakiutl face. The region surrounding
left fronto-malare is significantly smaller in
the average modified Kwakiutl than in the
average nonmodified Kwakiutl. There is a
significant 4% anterior-posterior increase
and a 3% to 6% medial-lateral decrease at
nasion, nasale, intradentale superior, and
the left premaxilla-maxilla junction. The 2%
anterior-posterior increase and 6% mediallateral decrease at the left maxillary tuberosity approaches significance. Other facial
landmarks show anterior-posterior increase
between 2% and 5% coupled with mediallateral and superior-inferior decreases.
Therefore, greater anterior-posterior growth
is associated with reduced medial-lateral
and superior-inferior growth in the face.
This pattern is consistent with changes observed in the cranial vault and cranial base.
The average face in the modified sample is
longer, narrower, and shorter than the average nonmodified face.
Nootka
Lateral and basal views of the average
nonmodified and modified Nootka are superimposed after generalized procrustes rotation in Figure 6. Differences between the
average nonmodified and modified Nootka
157
ANNULAR CRANIAL VAULT MODIFICATION
TABLE 3. Results of finite element scaling analyses of normal into modified Kwakiutl crania
Landmark
Size (Probl
Shape (Probl
0.89 (0.102)
2.80 (0.0241*
2.24 (0.0011"
-2.38 (0.9421
-1.25 (0.928)
1.58 (0.001Ix'
-0.19 (0.642)
-0.42 (0.744)
-2.02 (0.922)
-3.20 (0.999)"
-2.45 (0.994)"
-2.72 (0.9961"
-2.79 (0.980)*
-1.96 (0.9141
-0.72 (0.840)
-1.63 (0.982)"
7.40 (0.001)"
5.53 (0.0011"
7.13 (0.001)"
6.55 (0.0011"
3.80 (0.001)*
7.47 (0.001)"
3.41 (0.0011"
3.19 (0.001)"
4.84 (0.001)"
5.71 (0.0011"
7.91 (0.001)*
8.84 (0.001)"
2.85 (0.436)
2.04 (0.6281
5.21 (0.0011"
5.80 (0.0011"
0.54
-0.12
0.08
-0.35
-2.58
-0.53
-1.46
-1.10
5.68 (0.001)"
4.12 (0.0041"
5.36 (0.1821
4.98 (0.290)
2.32 (0.382)
2.39 (0.014)"
2.53 (0.278)
1.95 (0.470)
2.17 (0.644)
3.57 (0.278)
2.18 (0.5041
2.66 (0.326)
3.20 (0.006)"
2.72 (0.001)"
2.13 (0.024)"
2.97 (0.612)
2.93 (0.166)
3.65 (0.0561+
2.81 (0.144)
4.00 (0.001P
0.02 (0.0841+
PV1 (DIR)
PV2 (DIR)
PV3 (DIRI
~
Cranial Vault
Bregma
Bregma-lambda
Lambda
Lambda-opisthion
Opisthion
Pterion-aster. midpt.
Pterion (Rl
Pterion (Ll
Bregma-pterion (R)
Bregma-pterion (Lj
Bregma-asterion (Rl
Bregma-asterion (L)
Pterion-asterion (Rl
Pterion-asterion (L)
Asterion (Rl
Asterion (L)
Cranial Base
Jugular process (Rl
Jugular process (Li
Foramen lacerum (Rl
Foramen lacerum (Ll
Optic foramen midpt.
Vomer spine
Infratemporal Crest (Rl
Infratemporal Crest (Ll
Face
Fronto-malare (R)
Fronto-malare (Ll
Zygomaxillare sup. (R)
Zygomaxillare sup. (L)
Nasion
Nasale
Intradentale superior
Posterior nasal spine
Maxillary tuber. (Rj
Maxillary tuber. (Ll
Premaxilla-maxilla (Ri
Prernaxilla-maxilla (Lj
Global
Elements
Lower face (Rl
Lower face (L)
Upper Face (Ri
Upper Face (Ll
Cranial Base
Posterior vault (Rl
Posterior vault (Ll
Anterior vault (Ri
Anterior vault (L)
(0.338)
(0.5321
(0.472)
(0.558)
(0.984)*
(0.7301
(0.928)
(0.894)
-0.08 (0.548)
-2.43 (0.984)*
-0.59 (0.7261
-1.34 (0.884)
0.23 (0.3681
0.11 (0.448)
-0.73 (0.8241
0.72 (0.4101
-0.75 (0.7361
-1.84 (0.9361
1.01 (0.204)
-1.37 (0.9061
-0.74 (0.9461
Volume differences
-0.54 (0.606)
-2.12 (0.830)
-2.69 (0.902)+
-3.17 (0.9581+
-0.01 (0.500)
-1.64 (0.7601
-2.51 (0.8561
1.84 (0.120)
0.75 (0.3261
12.73 (A-PI
11.79 (A-PI
13.88 (A-PI
6.05 (A-P)
3.76 (A-PI
13.70 (A-PI
4.85 (A-PI
4.26 (A-PI
4.69 (AI-PSI
4.67 (AI-PSI
8.78 (AL-PM)
9.59 (AL-PM)
0.28 (A-Pi
0.55 (A-PI
7.10 (A-PI
6.89 (ALPMI
-3.79
-0.45
-1.66
-2.46
-1.70
-2.79
-2.32
-2.45
-3.28
-4.63
-4.28
-4.04
-2.01
-1.93
-4.12
-4.26
(0s-RIi
(S-I)
(M-L)
(M-L)
(M-L)
(M-L)
(S-I)
(MI-LSI
(M-L)
(MS-LII
(S-I)
(S-I)
(M-L)
(MS-LII
(M-Ll
(S-I)
8.95 (A-P)
5.87 (A-PI
8.22 (A-P)
7.04 (A-PI
0.44 (A-P)
2.90 (A-PI
1.79 (A-P)
1.66 (A-PI
-1.28
-1.76
-2.84
-2.70
-2.89
-1.84
-1.70
-2.13
(M-L)
(MS-LI)
(S-I1
(S-I1
(M-Ll
(S-I)
(M-Li
2.71 (S-I)
2.26 (MI-LSi
2.33 (AMI-PLS)
2.03 (AMI-PLSI
4.85 (A-PI
4.06 (A-PI
2.17 (A-P)
4.08 (AI-PSI
3.18 (A-PI
2.50 (A-PI
5.14 (A-P)
3.83 (A-P)
-0.21 (A-P)
-2.94 (MS-LI)
-1.03 (AMS-PLI)
-1.46 (AS-PI)
-1.04 (S-I)
-1.23 (0s-RI)
-1.29 (S-I)
1.49 (AS-PI)
-1.30 (MI-LSI
-1.46 (MI-LSI
-0.20 (M-Lj
-1.68 (S-I)
-
-
~
(S-I)
-4.52
-1.69
-3.69
-9.34
-5.34
-4.30
-2.75
-2.76
-6.65
-8.42
-9.84
-11.23
-6.17
-4.26
-4.31
-6.43
(01-RS)
(M-L)
(S-I)
(S-I)
(S-I)
(S-I)
(M-L)
(MS-LII
(AS-PI)
(MI-LS)
(AM-PL)
(AM-PLI
is-I)
(MI-LS)
(S-I)
(AM-PLj
-5 05 (S-I)
-3.96 (MI-LS)
-4.25 (M-L)
-4.63 (M-L)
-4.94 (S-I)
-2.48 (M-L)
-4.23 (S-I1
-2.68 (M-L)
-2.59
-6.08
-2.91
-4.33
-2.82
-2.28
-2.91
-3.13
-3.86
-6.08
-1.63
-5.73
(M-L)
(A-P)
(AL-PM)
(AMS-PLI)
(M-Ll
(RS-01)
(M-Ll
(M-Ll
(MS-LI)
(MS-LI)
(S-I)
(M-L)
Reported results include size and proportion of bootstrap samples exhibiting a greater size difference than that observed in the samples; shape
and the proportion of bootstrap samples exhibiting a greater shape difference than that observed in the samples; the principal values (PVil and
their associate anatomical directions (transverse directions for midsagittal landmarks are indicated as 0 for left and R for right); and the
proportional volume differences between normal and modified Kwakiutl for each finite element, and the proportion of bootstrap samples
exhibiting greater volume differences. Tests of local and global size and element volume are two-tailed tests. Landmarks or elements with
proportions >0.975 are significantly smaller than normal and proportions <0.025 are significantly larger than normal. Tests of local and global
shape change are one-tailed tests, and proportions <0.05 are significantly different from normal.
*Statistical significance a t P = 0.05 level.
+ Statistical significance a t P = 0 10 level.
are illustrated in Table 4 and Figure 7.
There are fewer significant differences between the nonmodified and modified
Nootka, but it is unclear if this is a product
of smaller sample size, or differences in
modification. These results are based on a
smaller sample (Nnm= 28, N, = 20) than
the analyses of the Kwakiutl (Nnm= 62,
N, = 40).
As with the Kwakiutl, the average modified male Nootka are 9% smaller than the
nonmodified individuals from the same
L.A.P. KOHN ET AL.
158
B
Fig. 5. Finite element scaling results of the deformation of average normal and modified Kwakiutl
drawn on the average normal Kwakiutl cranium. The ellipses represent the magnitudes and directions of
differences between normal and modified Kwakiutl. Ellipses are exaggerated by a factor of 2.5 to aid in
the interpretation of the effects of annular modification. The numerical results of this analysis are
presented in Table 3. A, lateral view, B, inferior view.
sample. The average nonmodified and modified Nootka females do not differ in volume.
All coordinates of the modified male Nootka
were resized to equalize the overall volume
of the nonmodified and modified Nootka
crania. After resizing, no significant global
size or shape difference between the nonmodified and modified individuals are observed.
Only two cranial vault landmarks exhibit
significant size differences between nonmodified and modified individuals. Modified
ANNULAR CRANIAL VAULT MODIFICATION
159
A
B
Fig. 6. Superimposition of average normal (solid line) and modified Nootka (dotted line) following
resizing and Procrustes rotation of all landmarks: A, lateral view, B, inferior view.
individuals are larger than nonmodified at
the midpoint between the pterion-asterion
landmarks, but smaller at the right bregmapterion. There are significant shape differences between nonmodified and modified
skulls at bregma, lambda, lambdaopisthion, the pterion-asterion midpoint,
right pterion, right and left pterion-asterion,
and right asterion. Shape change at the left
asterion approaches statistical significance.
The midline of the modified cranial vault is
approximately 5%longer along the anteriorposterior axis and is 5% smaller along the
superior-inferior axis. Little change occurs
along the medial-lateral axis in the midsagittal plane. The lateral cranial vault is increased 4% along an axis from anterior-latera1 to posterior-medial and decreased 5%
160
L.A.P. KOHN ET AL.
TABLE 4. Results of finite element scaling analyses of normal into modified Nootka crania
Landmark
Size (Prob)
Cranial Vault
Bregma
-0.82 (0.694)
Bregma-lambda
1.22 (0.268)
1.53 (0.064)
Lambda
-3.01 (0.942)
Lambda-opisthion
-0.29 (0.608)
Opisthion
Pterion-aster. midpt.
1.73 (0.012)"
Pterion (R)
0.21 (0.428)
Pterion (L)
-0.06 (0.508)
Bregma-pterion (R)
-1.48 (0.774)
Bregma-pterion IL)
-2.86 (0.932)
-2.55 (0.950)+
Bregma-asterion (R)
Bregma-asterion (L)
-2.22 (0.922)
-0.40 (0.584)
Pterion-asterion (Rj
Pterion-asterion (L)
-0.74 (0.626)
Asterion (R)
-1.62 (0.918)
Asterion (L)
-1.32 (0.862)
Cranial Base
Jugular process (R)
1.13 (0.246)
Jugular process (L)
-1.12 (0.668)
Foramen lacerum IR)
-1.32 (0.654)
Foramen lacerum (L)
2.31 (0.1781
Optic foramen rnidpt.
-2.02 (0.9161
Vomer spine
-0.11 (0.528)
Infratemporal crest (R) -0.66 (0.634)
Infraternporal crest iL)
-0.91 (0.666)
Face
Fronto-malare ( R )
1.32 iO.220)
Fronto-malare (L)
-0.64 (0.554)
Zygomaxillare sup. iR)
1.99 (0.192)
Zygomaxillare sup. iL)
1.93 (0.120)
Nasion
-0.01 (0.498)
Nasale
2.42 (0.032)+
Intradentale superior
1.52 (0.190)
Posterior nasal spine
-2.96 (0.922)
Maxillary tuber. (Rl
3.10 (0.0541
Maxillary tuber. (L)
1.62 (0.172)
Premaxilla-maxilla (R)
1.65 (0.224)
Premaxilla-maxilla (L)
1.17 (0.264)
-0.06 (0.522)
Global
Element
Volume difference
Lower face (R)
5.11 (0.138)
Lower face (L)
3.79 (0.156)
Upper face (R)
0.59 (0.404)
-1.22 (0.628)
Upper face (L)
Cranial base
2.37 (0.246)
Posterior vault (R)
-1.70 (0.694)
-1.79 (0.694)
Posterior vault (L)
Anterior vault (R)
0.35 (0.410)
-0.23 (0.4961
Anterior vault IL)
Shape (Prob)
4.53 (0.001)"
2.22 (0.478)
4.26 (0.001)*
6.24 (0.001)*
1.77 (0.514)
4.21 (0.001)*
2.11 (0.032)*
1.88 (0.062)
3.20 (0.112)
3.89 (0.1181
6.22 (0.012)*
6.69 l0.030)*
2.40 (0.852)
1.92 (0.882)
3.99 (0.032)*
4.01 (0.054)+
PV1 (DIR)
5.33 (A-P)
3.99 (AI-PSI
7.97 (A-P)
4.00 (A-P)
2.20 (A-P)
7.96 (A-P)
2.56 (A-P)
1.94 (A-PI
3.13 (ALI-PMS)
2.41 (ALI-PMS)
5.26 (AL-PM)
7.36 (AL-PM)
2.47 (S-I)
1.86 (S-I)
3.70 (AL-PM)
4.49 (AL-PMI
3.09 (0.456)
5.60 (A-P)
3.57 (A-PI
4.06 (0.218)
11.26 (A-P)
9.93 (0.1241
17.28 (A-P)
9.45 (0.124)
1.52 (M-L)
4.11 (0.062)+
1.98 (M-Ll
1.61 (0.4941
4.96 (0.088)+ 4.28 (MI-LS)
3.52 (0.306)
3.01 (MI-LS)
2.63 (0.784)
3.76 (0.382)
5.75 (0.110)
3.94 (0.452)
2.42 (0.360)
2.21 (0.264)
2.82 (0.038)*
5.62 (0.130)
3.65 10.300)
3.24 (0.404)
2.23 (0.652)
3.13 (0.328)
0.03 (0.352)
4.92 (S-I1
3.55 (S-I1
8.74 (M-L)
5.89 iM-L)
2.42 (AI-PSI
4.81 iM-L)
5.72 (AI-PSI
2.01 (A-P)
8.92 (S-I)
6.26 is-I)
5.02 (M-PSI
5.86 (AI-PS)
PV2 (DIR)
I
PV3 (DIR)
-1.56 (M-L)
1.43 (M-L)
-0.22 (M-L)
-1.59 (M-Ll
-1.01 (M-L)
0.51 (M-L)
0.74 (M-L)
0.52 (M-L)
-3.34 (MI-LS)
-3.79 (AS-PI)
-2.34 (S-I)
-4.31 (S-I)
-0.16 (M-Lj
-1.23 (M-Ll
-2.24 is-I)
-3.30 (AM-PLj
-5.61 (S-I)
-1.57 (AS-PI)
-2.53 (S-I)
10.08 (S-I)
-1.95 (S-I1
-2.63 (S-I)
-2.53 (S-I)
-2.52 (S-I)
-3.85 (AM-PL)
-6.53 (AMI-PLS)
-9.26 (AM-PL)
-8.26 (AM-PL)
-3.34 (A-P)
-2.73 (A-P)
-5.78 (AM-PL)
-4.61 (S-I)
0.16 (S-I)
-0.39 (MI-LSI
-0.05 (S-I)
0.06 (S-I)
0.22 6-1)
-0.28 (AS-PI)
1.43 (A-P)
-0.09 (A-P)
-2.04 (M-L)
-6.02 (MS-LII
12.27 (M-L)
-7.38 (M-L)
-7.21 (A-PI
- 1.94 (Mu-PSI
-6.97 (MS-LI)
-5.27 (MS-LII
1.00 (AM-PLI
0.30 (M-Ll
3.78 (AI-PS)
3.90 (AI-PSI
0.94 (M-L)
3.40 (AI-PSI
0.32 (M-L)
-0.02 (M-L)
1.16 (A-P)
0.93 (AL-PM)
0.40 (M-L)
-0.40 iM-L)
-1.69 iAL-PM)
-5.33 (A-P)
-5.42 (AS-PI)
-3.42 (AS-PI)
-3.23 (AS-PI)
-0.61 (AS-PI)
-1.16 (AS-PI)
--9.73 (S-I)
-0.06 (MS-LI)
-1.91 (AM-PLI
-0.23 (AS-PI)
- 1.60 (AS-PI)
'Reported results include size and the proportion of bootstrap samples exhibiting a greater size difference than that observed in the samples;
shape and the proportion of bootstrap samples exhibiting a greater shape difference than that observed in the samples; the principal values (PVi)
and their associate anatomical directions; and the proportional volume differences between normal and modified Nootka for each finite element.
and the proportion of bootstrap samples exhibiting greater volume differences. Tests of local and global size and element volume are two-tailed
tests. Landmarks or elements with proportions >0.975 are significantly smaller than normal and proportions <O 025 are significantly larger
Lhan normal. Tests of local and global shape change are one-tailed tests, and proportions <0.05 are significantly different from normal.
Statistical significance a t P = 0.05 level.
t Statistical significance a t P = 0 10 level.
along an axis from anterior-medial to posterior-lateral. Thus annular modification of
the Nootka cranial vault causes the cranial
vault to grow longer, narrower, and decreased in height. This assessment is complicated by the observation that the axes of
change are not entirely coincident with the
cardinal axes.
Modification of the cranial vault in the
Nootka produces few significant effects on
the cranial base. There are no size differences at cranial base landmarks, and shape
ANNULAR CRANIAL VAULT MODIFICATION
161
A
B
Fig. 7. Finite element scaling results of the deformation of average normal and modified Nootka drawn
on the average normal Nootka cranium. The ellipses represent the magnitudes and directions of differences between normal and modified Nootka. Ellipses are exaggerated by a factor of 2.5 to aid in the
interpretation the effects of annular modification. The numerical results of this analysis are presented in
Table 3. A, lateral view; B, inferior view.
change at the optic foramen midpoint and
the right infratemporal crest only approach
significance. There is a considerable amount
of variation observable in the cranial base.
This is evidenced by large shape changes at
the right and left foramen lacerum (11-16%
anterior-posterior increase and 7-12% medial-lateral decrease), both of which do not
approach significance. When both foramen
lacerum and the midpoint between the right
162
L.A.P. KOHN ET A L
and left optic foramina are excluded, the cranial base increases 3% anterior-posteriorly
and decreases the same amount medial-laterally and superior-inferiorly, i.e., the cranial base is slightly longer, narrower, and
shallower in the modified Nootka than in
the nonmodified Nootka. The average of the
small sample of Nootka modified crania can
be expected to be greatly influenced by extremal values in any of the individuals that
comprise the sample.
There are only two significant facial differences between nonmodified and modified
individuals. Size increase approaches significance at nasale. Intradentale superior exhibits a significant shape difference, with a
6% increase along anterior-inferior to posterior-superior axis and a slight anterior-superior to posterior-inferior decrease. In general, the modified face displays 5% increase
medial-lateral or anterior-inferior to posterior-superior and a 3% decrease anterior-superior to posterior-inferior.
DISCUSSION
The profound direct effects of annular
modification of the cranial vault are accompanied by indirect effects on the morphology
of the cranial base and face in both the
Kwakiutl and Nootka. In the Kwakiutl, cranial binding resulted in a reshaping of the
cranial vault. The medial-lateral and superior-inferior restriction of neurocranial
growth was compensated for by a large increase in anterior-posterior growth. Because
functional competence must be maintained
(i.e., the cranial vault, cranial base, and face
must remain articulated), increased anterior-posterior growth also results in localized shape changes in the cranial base and
face. As in the cranial vault, annular modification resulted in increased anterior-posterior and decreased medial-lateral and superior-inferior cranial base growth. Facial
growth also increased anterior-posteriorly
and decreased from anterior-lateral to posterior-medial.
Comparisons between Kwakiutl and
Nootka indicate that the Nootka cranial
vault is modified in the same direction as
was observed in the Kwakiutl. However,
there is a greater difference between the av-
erage nonmodified and modified Kwakiutl
cranial vault than was observed between
the average nonmodified and modified
Nootka cranial vault (see Tables 3 and 4).
Modification of the Nootka cranial base is
similar in direction and magnitude to that
observed in the Kwakiutl, although there
are fewer statistically significant effects of
modification in the Nootka. The directions
of modification of the Nootka face differ from
that observed in the Kwakiutl face, and
there are fewer significant effects of modification. These comparisons are reinforced
when the modified Kwakiutl and Nootka
crania are directly compared in Figure 8 after Procrustes rotation. This figure illustrates that modification in the Kwakiutl
crania appears to be more extreme than in
the Nootka, and this is consistent with ethnographic accounts in that cranial wrapping
materials in the Kwakiutl were frequently
changed and they may have exerted more
pressure than the deforming appliances of
the Nootka (Boas, 1921; Dingwall, 1931).
We can only speculate on the cause of the
differences in direction and magnitude of
modification between the Kwakiutl and
Nootka. We must caution that the sample
size of the Nootka is smaller than that of the
Kwakiutl and thus the results may be more
severely influenced by sampling error. With
a smaller sample size, i t is more difficult to
reach statistically significant results. Differences in the direction of facial growth in
the Kwakiutl and Nootka may result from
differences in modifying applicances in the
two groups.
The finite element scaling results in the
Kwakiutl are consistent with previous results and provide a better understanding of
the effect of annular modification. McNeill
and Newton (1965) and Anton (1989) noted
a n increase in cranial base angle (angle between nasion, sella, and basion) and a superior placement of the foramen magnum relative to the Frankfort Horizontal. Both
studies concluded that neurocranial landmarks and nasion were displaced posteriorsuperiorly relative to the Frankfort Horizontal resulting in the increase in cranial base
angle. Our finite element scaling results
demonstrate that, relative to our defined
ANNULAR CRANIAL VAULT MODIFICATION
163
I A
~
I
B
Fig. 8. Superimposition of average modified Kwakiutl (solid line) and modified Nootka (dotted line)
following Procrustes rotation of all landmarks: A, lateral view; B, inferior view.
axes, nasion is displaced along a vector from
anterior and inferior to posterior and superior. However, opisthion is also displaced superiorly relative to surrounding landmarks.
This results in a longer, shallower cranial
base and together with the displacement of
nasion explains the increased cranial base
angle observed in annularly modified crania.
Previous studies reported that modified
individuals exhibited a longer and narrower
cranial base and face (McNeill and Newton,
164
L.A.P.KOHN ET AL
1965; Cybulski, 1975; Anton, 1989). Our observations of increased anterior-posterior
growth and restricted medial-lateral and superior-inferior growth are consistent with
results from previous studies. There is a
consistent pattern of anterior-posterior increase in dimensions and an anterior-lateral
to posterior-medial decrease throughout the
Kwakiutl face. The extent of this morphological difference was not apparent in previous
studies because palate length was the only
anterior-posterior facial dimension included.
Comparison of the results of annular modification to anterior-posterior modification
suggest marked differences between these
types of modifications (see Cheverud et al.,
1992 for finite element scaling analyses of
anterior-posterior modification). Both studies used finite element scaling to describe
morphological changes in the neurocranium, cranial base and face. The cranial
vault modifications defining these two
classes are opposite, as are their effects on
the cranial base and face. Following Cheverud et al. (1992: Fig. 9), the effects of modification can be described by a simple geometric model, which depicts the cranial
vault as a quadrilateral and the face as a
triangle (Fig. 9). In the figure, the nonmodified cranial vault is represented by a quadrilateral, and the triangle represents the nonmodified face. The anterior wall of the
quadrilateral and the base of the triangle
represents the cranial base. Anterior-posterior modification produces medial-lateral
widening and an anterior to posterior shortening of the cranial vault. Because of these
modifications to the cranial vault, the cranial base is widened and the face must
shorten anterior-posteriorly to maintain
competence. In contrast, the annular modification results in a cranial vault that is
longer and narrower than nonmodified. The
cranial base is narrower, and the face is
longer and narrower than nonmodified. Opposite vault modifications produce opposite
facial effects in a predictable fashion. Shape
must be altered since overall size remains
unchanged.
We may use information from these culturally modified crania together with natural and experimental examples of cranial
modification to examine the growth relationships of the cranial vault, cranial base,
and face. Growth studies in animals with
experimentally manipulated crania (Young,
1959; Pucciarelli, 1978) or crania in which
sutures were manipulated to close prematurely (Babler and Persing, 1982; Babler et
al., 1987; Babler, 1988, 1989) reveal that
modification of growth patterns in the cranial vault are accompanied by changes in
morphology of the cranial base and face. For
example, Pucciarelli (1978) applied an annular band to the cranial vault of growing
rats. Indirect effects of annular cranial vault
modification were observed in the adult cranial base and face. In modified animals the
cranial base exhibited angular differences
from nonmodified and was narrower,
whereas the face was restricted in width and
height. We also found that annular modification caused restrictions in cranial base
and facial width. Babler (1989) reviews results from previous experiments in which
single or multiple cranial vault sutures were
closed in infant rabbits to observe the effect
of premature suture closure on the growing
cranial vault, cranial base, and face. Premature closure of the bilateral coronal, unilateral coronal, or frontonasal sutures produced predictable indirect morphological
effects on the cranial base and face. For example, bilateral coronal suture closure produced anterior-posterior shortening of the
cranial base and face, and flattened the angle of the cranial base (Babler, 1989). We
also found anterior-posterior shortening of
the cranial base and face in anterior-posterior modified crania (Cheverud et al., 1992).
Only closure of the sagittal suture had no
effect on cranial base or facial morphology
(Babler, 1989).This contrasts with our analyses of annular modification. Babler’s
(1989) experiments show that modification
of the single cranial vault suture will produce complementary and predictable effects
on the direction and magnitude of growth of
the cranial base and face. Premature closure
of multiple sutures (coronal, sagittal, and
interfrontal) is more complicated and also
produces indirect effects on the cranial base
and face; however, these three-dimensional
changes are less predictable than was seen
ANNULAR CRANIAL. VAULT MODIFICATION
165
NORMAL
ANTERO-POSTEERIORMODIFICATION
4
Fig. 9. Geometric representation of the effects of cranial vault modification on the morphology of the cranial
base and face. In this figure the cranium is represented
by a quadrilateral and the face is representedby a triangle. The cranial base is represented by the base of the
triangle and the anterior wall of the quadrilateral. As
shown by Cheverud et al. (19921, anterior-posterior
modification produces a shortening of the cranium along
the anterior to posterior dimensions. The cranial base is
widened along the medial-lateral dimension. The face
must become shorter and wider in order to articulate
with the cranial vault and base. In contrast, annular
modification results in a lengthening of the cranium
along the anterior-posterior dimension, and a decrease
along the medial-lateral dimension. The cranial base is
narrower, and the face is longer and narrower than normal. The normal and modified crania are superimposed
and can be directly compared in the bottom figure. In
both types of modification, modification of the face is
required for conformation of the face with the cranial
base.
with premature closure of single sutures
(Babler, 1988).
Individuals with naturally occurring premature closure of cranial vault sutures, or
craniosynostosis, also exhibit abnormal
growth patterns in the cranial vault, cranial
base, and face when compared to nonmodified individuals (Kreiborg, 1986; David,
1989; Striker et al., 1990). Individuals with
oxycephaly (bilateral coronal synostosis)
display anterior-posterior reductions of the
cranial vault, cranial base, and maxilla
(Kreiborg, 1986). This compliments our results of anterior-posteriorly modified crania
(Cheverud et al., 1992). Plagiocephaly (unilateral coronal synostosis) is characterized
by marked asymmetry of the cranial vault,
cranial base, and face, and shortening of the
anterior cranial base (Kreiborg, 1986). Apert and Crouzon syndromes, which include
premature synostosis of multiple cranial
vault sutures (coronal, sagittal and lamb-
166
L.A.P. KOHN ET AL
doid), differ from nonmodified in numerous
cranial base and facial dimensions (Richtsmeier, 1985,1987, 1988; Kreiborg, 1986).
It has been hypothesized that one cause of
craniosynostosis is premature closure of the
sphenoethmoid synchrondrosis (Burdi,
1986) or cranial base sutures (Moss, 1959,
1975; Tessier, 1971). Other factors have also
been proposed to cause craniosynostosis
(Cohen, 1986; Graham et al., 1979). The observed patterns of growth are independent
of whether the primary abnormality is in the
cranial vault or in the cranial base. Results
of natural and experimental growth studies
indicate that there is a developmental interdependence of components of the cranial
vault with the cranial base and face (Enlow,
1990). Redirection of growth vectors in the
cranial vault produce correlated responses
in the other functional components (Moss,
1958). Such correlated growth responses are
adaptive since the cranial base and face
must articulate with and function with the
cranial vault. Moss and Young (1960) suggested that modifications of growth,
whether internally (e.g., premature suture
closure) or externally derived (e.g., cultural
modification), produce growth forces that
are secondary to nonmodified growth forces.
We thus conclude that modification of
growth patterns of the cranial vault can be
expected to influence the morphology of the
cranial base and face. Variation in growth
patterns produces variation in adult morphology.
ACKNOWLEDGMENTS
We thank Dr. Glenn Cole and the Field
Museum of Natural History for access to
their skeleton collection, and Ms. Jaymie L.
Brauer and the American Museum of Natural History for access to the Nootka and
Kwakiutl series. We also thank Dr. Lyle W.
Konigsberg for his valuable help in the initial stages of this research and for the photograph of Field Museum specimen number
40797, and Dr. Joan T. Richtsmeier for making FIESCA available to us. Nyuta Yamashita and Dr. Jim Midkiff helped in data collection. Morgan Robertson assisted with data
analysis. Three anonymous reviewers provided valuable comments on the manu-
script. This research was supported by NSF
grant BNS 89-10998 to J.M.C.
LITERATURE CITED
Allison M, Gerszten E, Munizaga J, Santoro C, and Focacci G (1981)La practica de la deformacion craneana
entre 10s pueblos Andinos Precolombinos. Chugara
7t238-260.
Anton SC (1989) Intentional cranial vault deformation
and induced changes of the cranial base and face. Am.
J. Phys. Anthropol. 79:253-268.
Babler WJ (1988) Effects of multiple suture closure on
craniofacial growth in rabbits. In KWL Vig and AR
Burdi (eds.): Craniofacial Morphogenesis and Dysmorphogenesis. Ann Arbor: Center for Human
Growth and Development, pp. 73-90.
Babler WJ (1989) Relationship of altered cranial suture
growth to the cranial base and midface. In JA Persing,
MT Edgerton and JA Jane (eds.): Scientific Foundations and Surgical Treatment of Craniosynostosis.
Baltimore: Williams and Wilkins, pp. 87-95.
Babler WJ, and Persing JA (1982) Experimental alteration of cranial suture growth: Effects on the neurocranium, basicranium, and midface. In AD Dixon and
BG Sarnat (eds.):Factors and Mechanisms Influencing Bone Growth. New York: Alan R. Liss, pp. 333345.
Babler WJ, JA Persing, WJ Nagorsky, and JA Jane
(1987) Restricted growth at the frontonasal suture:
Alterations in craniofacial growth in rabbits. Am. J .
Anat. 178:90-98.
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. 85t173-192.
Boas F (1921) Ethnology of the Kwakiutl: based on data
collected by George Hunt. 35th Ann. Rep. Bur. Am.
Ethnol.:39-794.
Bookstein FL (1978) The Measurement of Biological
Shape and Shape Change. Lecture Notes in Biomathematics, No. 24. New York: Springer-Verlag.
Bookstein FL (1983) The geometry of craniofacial
growth invariants. Am. J . Orthod. 83t221-234.
Boodstein FL (1984) A statistical method for biological
shape change. J. Theor. Biol. 107:475-520.
Bookstein FL (1987) Describing a craniofacial anomaly:
Finite elements and the biometrics of landmark locations. Am. J. Phys. Anthropol. 74:495-510.
Brothwell D (1975) Possible evidence of a cultural practice affecting head growth in some late Pleistocene
East African and Australian populations. J. Arch. Sci.
2: 75-77.
Burdi AR, Kusnetz AB, Venes JL, and Gebarski SS
(1986) The natural history and pathogenesis of the
cranial coronal ring articulations: Implications in understanding pathogenesis of the Crouzon craniostenotic defects. Cleft Palate J . 23t28-39.
Cheverud JM, Kohn LAP, Konigsberg LW, and Leigh SR
(1992) The effects of fronto-occipital artificial cranial
ANNULAR CRANIAL VAULT MODIFICATION
vault modification on the cranial base and face. Am. J.
Phys. Anthropol. 88:323-345.
Cheverud JM, Lewis JL, Bachrach W, and Lew WB
(1983) The measurement of form and variation in
form: An application of three-dimensional quantitative morphology by finite element scaling methods.
Am. J. Phys. Anthropol. 62t151-165.
Cheverud JM, and Midkiff J E (1992) Effects of frontooccipital reshaping on mandibular form. Am. J. Phys.
Anthropol. 87:167-171.
Cheverud JM, and Richtsmeier J T (1986) Finite-elernent scaling applied to sexual dimorphism in rhesus
macaque (Macaca mulattai facial growth. Syst. Zool.
35r381-399.
Cole D (1985) Captured Heritage: The Scramble for
Northwest Coast Artifacts. Vancouver, BC: Douglas
and McIntyre.
Cohen MM (1986) The etiology of craniosynostosis. In
MM Cohen (ed): Craniosynostoses: Diagnosis, Evaluation, and Management. New York: Raven Press, pp.
59-79.
Cybulski JS (1975) Skeletal variability in British Columbia coastal populations: a descriptive and comparative assessment of cranial morphology. Nat. Mus.
Can., Nat. Mus. Man, Mercury Ser., Archaeol. Survey
Can., pap. 30.
David DJ, Hemmy DC, and Cooter RD (1989) Craniofacia1 Deformities. New York: Springer Verlag.
Dingwall J E (1931) Artificial Cranial Deformation: A
Contribution to the Study of Ethnic Mutilations. London: Bale and Danielsson.
Dorsey GA (1895) Crania from the necropolis of Ancon,
Peru. Proc. Amer. Assoc. Adv. Sci. 43t358-370.
Dorsey GA (1897) Wormian bones in artificially deformed Kwakiutl crania. Am. Anthropol. 10t169-173.
Enlow DH (1990) Facial Growth, 3rd ed. Philadelphia:
Saunders.
Goodall C, and Bose A (1987) Models and procrustes
methods for the analysis of shape differences. Proceedings of the lgth Symposium of the Interface between Computer Science and Statistics, pp. 86-92.
Graham JM, deSaxe M, and Smith DW (1979) Sagittal
craniosynostosis: Fetal head constraint as one possible cause. J . Pediatr. 95:747-750.
Hamner RS, and Bachrach WE (1986) SCALE. Northwestern University Rehabilitation Engineering Program, Chicago.
Hrdlicka A (1905) Head deformation among the Klamath. Am. Anthropol. 7t360-361.
Konigsberg LW, Kohn LAP, and Cheverud JM (1992)
Cranial deformation and nonmetric trait variation (in
press).
Kreiborg S (1986) Postnatal growth and development of
the craniofacial complex in premature craniosynostosis. In MM Cohen (ed.): Craniosynostosis: Diagnosis,
Evaluation, and Management. New York: Raven
Press, pp. 157-189.
Lele S, and Richtsmeier J T (1991) Euclidian Distance
matrix analysis: A coordinate-free approach for comparing biological shapes using landmark data. Am. J.
Phys. Anthropol. 86t415-429.
Lewis JL, Lew WB, and Zimmerman J L (1980)A nonho-
167
mogeneous anthropometric scaling method based on
finite element principles. J . Biomech. 13t815-824.
Lozanoff S, and Diewert V (1986)Measuring histological
form change with finite element methods: An application using diazo-0x0-norleucine (DON) treated rats.
Am. J. Anat. 177:187-201.
McNeill RW, and Newton GN (1965) Cranial base morphology in association with cranial vault deformation.
Am. J. Phys. Anthropol. 23t241-254.
Morris GR (1989) FIESCA: Graphic and analytical software for finite element scaling analysis in biological
research. Johns Hopkins University, Department of
Civil Engineering.
Morton S (1839) Crania Americana. Philadelphia: Pennington.
Moss ML (1958) The pathogenesis of artificial cranial
deformation. Am. J. Phys. Anthropol. 16:269-286.
Moss ML (1959)The pathogenesis of premature cranial
synostosis in man. Acta Anat. 37r351-370.
Moss ML (1975)Functional anatomy of cranial synostosis. Child's Brain 1:22-33.
Moss ML (1988) Finite element method comparison of
murine mandible form differences. J. Cranio. Gen.
Dev. Biol. 8t3-20.
Moss ML, Skalak R, Pate1 H, Sen K, Moss-Salentijn L,
Shinozuka M, and Vilmann H (1985) Finite element
method modeling of craniofacial growth. Am, J.
Orthod. 87r453-472.
Moss ML, and Young RW (1960)A functional approach
to craniology. Am. J. Phys. Anthropol. 18t281-292.
Moyers RE, and Bookstein FL (1979)The inappropriateness of conventional cephalometrics. Am. J. Orthod.
87t453-472.
Oetteking B (1924) Declination of the pars basilaris in
normal and artificially deformed skulls: A study
based on skulls of the Chumash of San Miguel Island,
California, and those of the Chinook. Indian Notes
and Monographs 7. New York: Heye Foundation.
Oetteking B (1930) Craniology of the North Pacific
Coast. Mem. AMNH 16111:1-391.
Pucciarelli HM (1978) The influence of experimental
deformation on craniofacial development in rats. Am.
J . Phys. Anthropol. 48r455-461.
Richtsmeier J T (1985) A study of normal and pathological craniofacial morphology and growth using finite
element methods. Ph.D thesis, Department of Anthropology, Northwestern University.
Richtsmeier J T (1987) Comparative study of normal,
crouzon, and apert craniofacial morphology using finite element scaling analysis. Am. J. Phys. Anthropol.
74t473-493.
Richtsmeier J T (1988)Craniofacial growth in apert syndrome as measured by finite-element scaling analysis. Acta Anat. 133:50-56.
Richtsmeier JT, and Cheverud J M (1986)Finite element
scaling analysis of human craniofacial growth. J.
Craniofac. Genet. Dev. Biol. 6t289-323.
Rothhammer F, Cocilovo JA, Quevedo S, and Llop E
(1982) Microevolution in prehistoric Andean populations: I. Chronologic craniometric variation. Am. J.
Phys. Anthropol. 58:391-396.
168
L.A.P. KOHN ET AL
Sapir E (1922) Sayach‘apis, a Nootka trader. In EC Parsons (ed.):American Indian Life. New York: Huebsch.
Schendel SA, Walker G, and Kamisugi A (1980) Hawaiian craniofacial morphometrics: average Mokapuan
skull, artificial cranial deformation, and the “rocker”
mandible. Am. J. Phys. Anthropol. 52r491-500.
Striker M, Van Der Meulen J, Raphael B, and Mazzola B
f 1990) Craniofacial malformations. Churchill Livingstone: New York.
Tessier P (1971) Relationship of craniosysostoses to
craniofacial dysostoses, and to faciostenoses. Plast.
Reconstr. Surg. 48:224-237.
Young RW (1959) The influence of cranial contents on
postnatal growth of the skull in the rat. Am. J. Anat.
lO5:383415.
Документ
Категория
Без категории
Просмотров
0
Размер файла
1 547 Кб
Теги
base, effect, cranial, facer, vault, modification, annular
1/--страниц
Пожаловаться на содержимое документа