Hambly 1940 Craniometry of New Guinea

W2.
05
CRANIOMETRY OF NEW GUINEA
BY
WILFRID D. HAMBLY
CURATOR, AFRICAN ETHNOLOGY
THE LIBRARY OF THE
.31940
UNIVERSITY Of ILLINOIS
'
•WniiMrtmmnpir JFIIP imnniiirnnnomni
^/CAGO
ANTHROPOLOGICAL SERIES
FIELD MUSEUM OF NATURAL HISTORY
VOLUME XXV, NUMBER 3
FEBRUARY 29, 1940
PUBLICATION 465
THE UCBAHf
OF THE
UNIVERSITY Itf lUiMI
138 14-2 144 14-6
h
200 Kilometers
About- 125 Miles
MADANG
Former
FRIEDRICH
WILHELM_|
HAFEN
10
CENTRAL NEW GUINEA

For explanation of numbers see tables, pages 88, 89
BY
WILFRID D. HAMBLY
CURATOR, AFRICAN ETHNOLOGY
THE LIBRARY OF THE
MAR 18 1940
UNIVERSITY OF ILLINOIS
ANTHROPOLOGICAL SERIES
FIELD MUSEUM OF NATURAL HISTORY
VOLUME XXV, NUMBER 3
FEBRUARY 29, 1940
PUBLICATION 465
TKINTED IN THE UNITED STATES OF AMERICA
BY FIELD MUSEUM PRESS
Fft
v;26
3 CONTENTS
PAGE
List op Illustrations 85
I. The Nature and Provenance of the Field Museum Collection . . 87
II. Technique of Measurement 90

General Principles 90
Remeasurement of Fifty Skulls 92
III. Age and Sex in the Collection 122
IV. Grouping of the Data 126

V. Frequency Distributions and Sex Differences 222
Introduction 222
Interpretation of Diagrams 222
Measurements Made in situ near Merauke 225
VI. Contours of Skulls from New Guinea and Other Regions . 239
. .
VII. Comparative Study of Craniometry of New Guinea and Adjacent

Regions 244
Introduction 244
The Coefficient of Racial Likeness 244
Comparison of Average Measurements 263
Conclusions 270
Appendix: Tables Giving Coefficients of Racial Likeness . . . 273
Bibliography 282
Index 287
OS
83
LIST OF ILLUSTRATIONS
PLATES
XXXI. Male (left) and female of Borbor.
XXXII. A male of Watam, Kirau, front and side views.
XXXIII. A male of Sissano.
XXXIV. Two females of Merauke, front and side views.
XXXV. A
female of Merauke, front and back views.
XXXVI. Head-hunting trophies of Merauke, photographed in situ.
XXXVII. Two warriors of Merauke, and two newly acquired heads from a
village west of Merauke.
XXXVIII. A male of Merauke, front and side views.
XXXIX-LXXIV. Skulls of various types that have been measured. Identifica-
tion numbers are given in the legends. Reference should be made
to pages 130-221 for details of measurement.
TEXT FIGURES PAGE

12. Cranial capacities 226
13. Lengths of skulls 227
14. Breadths of skulls 228
15. Cranial indices 229
16. Skull heights, basion-bregma 230
17. Horizontal arcs over glabella 231
18. Horizontal arcs over ophryon 232
19. Sagittal arcs 233
20. Vertical arcs over bregma 234
21. Bizygomatic widths 235
22. Nasal indices 236
23. Facial indices 237
24. Facial triangles 238
25. Baining males, Gazelle Peninsula, New Britain, front and side views . . 255
26. Melanesian male, Gazelle Peninsula, New Britain, front and side views . 257
Map of New Guinea

MAPS PA
S 81
DRAWINGS
Average Contour Drawings of Skulls of New Guinea (natural size)
Vertical 238
Horizontal 238
Sagittal 238
Composite Contour Drawings of Skulls of New Guinea (natural size)

Vertical 240
Horizontal 240
Sagittal 240
Average Contour Drawings of Skulls of New Guinea Compared with Those of

New Britain, Australia, and the Wateita of East Africa (natural size)
Vertical 242
Horizontal 242
Sagittal 242
85
I. THE NATURE AND PROVENANCE OF THE
FIELD MUSEUM COLLECTION
The material studied in this publication was acquired mainly
by the Joseph N. Field Anthropological Expedition to the South
Sea Islands, 1909-13, under the leadership of Dr. A. B. Lewis,
Curator of Melanesian Ethnology, to whom I am indebted for the
explanations and notes given in the tables (pp. 88, 89, 223), also for
some photographs of physical types (Plates XXXI-XXXVIII,
Figs. 25, 26) and for some measurements made in situ near Merauke.
A smaller number of skulls was collected by Curator G. A.
Dorsey, who visited New Guinea in 1908. A few purchases were
made from J. F. G. Umlauff, who obtained the material from Cap-
tain H. Voogdt. Some of the skulls were obtained from F. Boas,
W. D. Webster, and J. Nathan. The only research work on this
collection is that of G. A. Dorsey (1897), who made a few measure-
ments on fifteen skulls.
Students interested in the general geography and ethnology of

New Guinea a well-written article in the Encyclopaedia
will find
Britannica (14th ed., vol. 16). A publication prepared by Dr. A. B.
Lewis (1932) gives an entertaining and instructive account of the
Melanesians. Seligman (1909) has given a helpful classification
of the natives of British New Guinea.
The general provenance of the material is fairly well established,
but by no means can we be certain that each skull belonged to an
inhabitant of the village from which the skull was obtained. The
Merauke people are head-hunters; so also are those of Strickland
River,and probably the inhabitants of Girara also. Skulls from
the Purari Delta probably combine head-hunters' trophies and
ancestral skulls. Skulls from Borbor, Murik, Kirau, Bure, Sissano,
Warapu, and Dallmannhafen are probably ancestral (see Map).
Yet the material, though inadequate for the study of family
or village groups, is, in the aggregate, sufficient for the determination
of the craniometrical characters of males and females of two main
localities; namely, portions of the northern and southern coasts
of the island of New Guinea.
In the first place, measurements and statistical work took cog-
nizance of the local groups however small, and of the two main
topographical regions, northern and southern. Then statistical
87
88 Craniometry of New Guinea
methods were used for determining the significance of the group

with a view to ascertaining with what validity the
differences,
groups might be combined and considered as one population.
The general scope of the research aims at:
(1) Studying vital statistics by noting the ages of all skulls

measured.
(2) Establishing from the data the relative craniometrical
characteristics of males and females. Forty-four charac-
ters have been measured, and, from these, fifteen indices have
Provenance of Skulls from Northern New Guinea

Field Museum Collection 89
been worked out. The study includes a comparison of small

local groups, and of northern and southern populations.
(3) Instituting comparisons with other craniometric measure-
ments from New Guinea and other areas.
(4) Testing the value of formulae for estimating cranial capacity.
This can be done by comparing the results obtained by use
of formulae with the figures resulting from direct measure-
ment by the mustard seed method.
(5) Remeasuring fifty skulls as a test of accuracy in technique.
The work was carried out in the Laboratory of Physical Anthro-
pology, Field Museum Natural History. For advice on points
of
of technique and method thanks are due to Dr. Gerhardt
statistical
von Bonin, who generously co-operated. Dr. G. K. Neumann,
of the University of Chicago, kindly assisted in making 360 contour
drawings.
Provenance of Skulls from Southern New Guinea

II. TECHNIQUE OF MEASUREMENT
General Principles
A student who wishes to acquaint himself with the difficulties
of technique will find that the literature falls into two categories.
On the one hand there are general historical surveys dealing with
anthropometric measurements since the time of Paul Broca, about
eighty years ago; and on the other hand many papers have been
devoted to discussion of the details of craniometrical and other
techniques. From the long list of articles available I would specially
recommend those Miss M. L. Tildesley (1938) since these give a
of
growth of the problem, together with
clear historical outline of the
a summary of the proposed work of the Comite de Standardisation
de la Technique Anthropologique. Other useful summaries are
those of Morant, Tildesley, and Buxton (1932), and Howells (1937b).
For my own work I have relied chiefly on the definitions of the
London Biometric School, whose technique is well summarized by
Morant (1927, p. 417; 1928, p. 362). Morant clearly indicates the
degree of agreement between his definitions and those of R. Martin
(1928). At certain points I shall refer to Martin's technique and
to his illustrations of instruments, but one cannot give a full account
of Martin's technique here, since his definitions, explanations, and
diagrams occupy a hundred pages. Some of the troublesome points
in Martin's technique and definitions are discussed by Wagner
(1937, pp. 32-38).
have derived great help from an article published by Buxton
I
and Morant (1933), and if a student is in doubt about the technique

I have followed he will find the explanation in that article. Buxton
and Morant give clear photographs illustrating their definitions of
porion, nasion, lambda, opisthion, basion, and the nasal spine. My
method of making measurements on the palate is that shown by
Buxton and Morant (1933, Plate XII, Fig. C).
The following summary of points of measurement and definitions
is given for ready reference. To aid comparative study the measure-

ments include three methods of taking nasal height, three methods
of measuring the orbits, and two methods of recording the length
of the palate. Notes on the correlation of my H' and BQ' with
other methods of recording the height of the skull and the trans-
verse arc will be found in an article by G. M. Morant (1928, p. 362).
90
Technique of Measurement 91
List of Measurements and Indices Taken on the Skulls of New Guinea

L= Length, maximum glabello-occipi- inner alveolar walls of the second
tal, in the median sagittal plane molars
B = Breadth, maximum calvarial Gi'= Shorter length of the palate
B'= Minimum frontal breadth between excluding the spine, measured
the temporal lines from the base of the posterior spine
H'= Height, basion-bregma EH = Height of the palate, taken with
LB = Length of base, basion-nasion Pearson's uraniscometer in the
Si = Frontal arc, nasion-bregma median line between the second
S/= Frontal chord, nasion-bregma molars. The measurements were
82= Parietal arc, bregma-lambda repeated with a more accurate in-
S2 ,= Parietal chord, bregma-lambda strument supplied by P. Hermann
S3 = Occipital arc, lambda-opisthion of Zurich
S 3 '= Occipital chord, lambda-opisthion f ml = Length of the foramen magnum
S = Sagittal arc, nasion-opisthion from opisthion to basion

GLU = Horizontal arc through glabella f mb = Breadth of the foramen magnum
and the most prominent part of 100 B/L = Cranial index

the occipital region 100 H'/L= Height-length index
U = Horizontal arc through ophryon 100 B/H'= Breadth-height index
and the most prominent part of 100 (B-H') /L = Breadth -height -length
the occipital region index
BQ'= Vertical arc over bregma from OcI=S3/S 3 '= Occipital index (see M. L.
porion to porion Tildesley, 1921)
GB = Breadth of face between the 100 G'H/GB = Facial index
lowest points of the maxillo- 100 NB/NHR=Nasal index
zygomatic sutures 100 NB/NH'= Nasal index
J = Maximum bizygomatic breadth 100 O2/O1 = Orbital index, right and left
G'H = Height of upper face, nasion to 100 2 /Oi'= Orbital index, right
alveolar point 100 2 /Lac Oi = Orbital index, right
GL =Basion to alveolar point 100 G 2 /Gi= Palatal index
NHR and NHL = Nasal height, right 100 G 2 /Gi'= Palatal index
and left. Nasion to the lowest 100 EH/G 2 = Palatal height-breadth
edge of the pyriform aperture on index
the right orleft side 100 fmb/f ml = Index of the foramen
NH'= Nasal height, nasion to the base magnum
of the anterior nasal spine A Z = Alveolar angle
NB = Greatest breadth of the pyriform B Z= Basion angle
aperture N Z= Nasion angle
OiR = Breadth of right orbit, from the
P Z= Profile angle
point where the naso-frontal suture
Pros P Z. The angles A, B, N were
is cut by prolongation of the orbital
measured with a protractor from
margin the facial triangle which was con-
2 R = Height of the right orbit structed from the measurements
OiL=Breadth of the left orbit LB, GL, and G'H. The alveolar
2 L = Height of the left orbit P angle is the angle which the
Oi'(R and L) = Length of the orbit, right Frankfurt horizontal plane forms
and left, measured from the point with a line joining nasion and
where the naso-lachrymal suture alveolar point. This is the Frank-
meets the naso-frontal suture furt profile angle. The alveolar
Lac OiR and L = Right and left, from point is the lowest point of the
the point where the lachrymal crest alveolar margin, between the two
meets the fronto-lachrymal suture upper central incisors. The pros-
SC =Interorbital width from dacryon thionPZ is formed by the Frank-
to dacryon furt horizontal plane and the
Gi = Length of the palate, from the tip nasion-prosthion line. Prosthion
of the posterior nasal spine to the is the foremost point in the mid-
intersection of the median sagittal sagittal line. The instruments
plane and an imaginary line tan- used for measuring the PZ and the
gential to the inner alveolar margin prosthion PZ were Mollison's
of the middle incisors craniophore and the stativgoni-
G; = Breadth of the palate between the ometer
Remeasurement of Fifty Skulls
This experiment was made at the suggestion of Miss M. L.
Tildesley during her visit to Chicago. A knowledge of the errors
likely when measurements are retaken by the original

to occur
worker, using the same technique as that employed during first
measurement, is essential to the standardizing of technique. It
is also desirable that the same series should be remeasured by an-

other worker.
The designation numbers of the fifty male skulls selected at
random for remeasurement, six months after the first measurement,
are given in a table (pp. 98-121). Averages are given (pp. 116-121)
for the first and second series of measurements, also the AP/ A values
for each pair of averages.
—
Cranial Capacities. The technique for measuring cranial capac-
itiesdeserves particular study, since the subject has been one for
argument since the time of P. Broca. A review of the methods
employed by various workers has been given by Macdonell (1904),
and a detailed criticism of technique has recently been published
by Stewart (1937).
For measuring cranial capacity I made three preliminary ex-
periments with a crane etalon which was found to have a capacity
for 1,315 cc. of water. The mustard seed needed to fill this skull
weighed 1,026.7 gms.; this average resulted from three fillings.
Therefore, we obtain the ratio 1.281 to express the relation between
weight of seed and cubic centimeters of water. Dr. G. von Bonin
(1934) gives a list of such ratios obtained in eight experiments per-
formed by six different observers. On the average these observers
found that the ratio 1.284 expresses the relation between weight of
seed and cubic centimeters of water.
Macdonell (1904) found that the discrepancy in average meas-
urements for some fairly long series, measured by different workers
who used the same technique, was likely to be 10 to 15 cc. Von Bonin
(1934) concluded that the personal equation in measuring the aver-
age of a series may vary from 10 to 20 cc. Todd (1923) thought
that for the measurement of any individual skull the personal equa-
tion might be as much as 40 cc. Wagner (1937, p. 38) states that,
owing to different techniques, "it is best to disregard all average
differences under 30-40 cc, for data given by different authors,
and even this figure is sometimes too small."
I believe that, so far as the personal error is concerned, there is a
possibility of using the technique of filling with mustard seed more

accurately than most observers have believed. The use of sand or

fine shot (No. 8) gives a result which is in excess of the true capac-
ity as measured by water. Seed gives a result which approximates
most closely to that given by water.
My experiment showed the average capacity of fifty skulls
first
to be 1,258 cc, and the second experiment gave the almost identical
result of 1,257 cc. This indicates that the mustard-seed method is
reliable for obtaining the average capacity of a fairly long series;
neither can the measurement of individual skulls be considered
unsatisfactory.
There were three instances of identical measurements of capacity,
and 23 instances of differences of 5 cc. or less. For half the series,
therefore, the error is very small. The greatest error was —30 cc.
for skull No. 43502, but this is due to the fact that the base around
the foramen magnum was broken away, and at best the measure-
ment could be only an approximation. The average error is — 1 .33 cc.
per skull.
To get such accuracy half an hour must be spent in filling and
tamping the seed in each skull, and the error in weighing the seed
must be not greater than 1 gr. It is true that an approximate capac-
ity of ±20 to 30 cc. can be obtained by spending only fifteen minutes
in filling a skull and weighing the seed, but great patience is needed
to get results diverging not more than ±5 cc. for an individual
skull. The skull may seem to be full after ten minutes of filling
and tamping, but in the last stages of filling, a small spoon should
be used for feeding seed into the foramen magnum. If this is done,
with constant tapping, about 20 to 30 cc. of seed can be introduced
after one might think that the skull was really filled.
Before leaving the subject of technique there is a further point

of importance to consider. Filling a skull to full capacity with
seed takes from twenty to thirty minutes, and the process of tamp-
ing is extremely tedious. But, as previously explained, a second
tamping of seed, which is poured from the skull into a measuring
glass, can be avoided by multiplying the weight of the seed by a
constant factor, and so calculating the volume. Yet some workers
have imposed on themselves the task of measuring the seed in a
graduated vessel, and there can be no doubt that the tedium of
the experiment leads to a loose filling of the graduated cylinder,
and consequently the volume recorded is too high.
I am convinced that more research should be concentrated on
developing a quick technique for measuring the capacity of skulls.

I have tried measuring the sounder skulls by filling them with water,
which was run from a graduated vessel. But, generally speaking,

so much time is spent in plugging skulls with wax or plasticine in
order to make them water-tight, that one might as well follow the
method of filling with seed.
Calculating Cranial Capacities. — The best known of the formulae

for calculating cranial capacities are:
(1) Isserlis' formula, based on measurement of male Negro

skulls (1914, p. 189).
C= .0003849 X BLH + 96 ± 65/VN"
(2) Lee's general formula, based on the measurement of male

skulls from Germany, the Ainu, and Naqada (Lewenz and
Pearson, 1904, p. 390).
C= .000337 X HLB + 406.01
(3) Hooke's (1926, p. 33) formula, based on a study of male

English skulls.
C= .000366 X LBH' + 198.9 ± 45.8/VN"
(4) Bonin's (1934, p. 14) formula for estimating the capacity

of skulls from New Britain.
C=. 000263 X BLH + 404.9 ±35.1/VN"
(5) Manouvrier's formula as used by van den Broek (1923a,

vol. 7, p. 172).
C = LBH/2.4 for males
The measured capacity of 124 male crania from New Guinea

was 1280 ±6.87 cc. The probable error was calculated by using
the standard deviation for a long Egyptian series (Pearson and
Davin, 1924).
The table below indicates which formulae will give a cranial
capacity nearest to that actually measured with mustard seed.
Table of Calculated Cranial Capacities

and 1 cc, respectively, between measured and calculated means,

could well be used for estimating the capacity of any New Guinea
series of male skulls which the average L, B, and H' came close
in
to Field Museum Lee's formula gives the greatest delta
series.
value, and would not be serviceable for estimating the average
cranial capacity of male New Guinea skulls.
Hooke's formula gives a rough approximation to the measured

capacity. Bonin's formula for skulls from New Britain could not
be satisfactorily applied to our sample from New Guinea. We need
not be surprised that the formula of Isserlis gives such a close approxi-
mation to measured capacity. The formula was worked out on
Negro skulls from Africa, and, as we will show later, there is statisti-
cal evidence for a Negro element in our New Guinea sample.
Let us now apply the formula of Isserlis to estimate the capacities

of furthersamples of skulls from New Guinea and Africa.
Cranial Capacities by Formula of Isserlis

and by Direct Measurement
Region Observer
(pp. 116-121) we see that L, B, and B' have small differences (+0.5,
+0.3, —0.1 mm., respectively). The difference in measurement of H'
is zero. All the arcs and chords, S to S/, have differences in aver-
t
age measurement of 1.0 mm. or less.
The remeasurement of facial angles may be regarded as satis-

factory, though the differences in averages are slightly greater than
in case of linear measurements. The greatest discrepancy is —1.3°
for the N angle. It is gratifying that the differences in averages
for the P and for the prosthion P angles are only —0.7°. The skull
was a craniophore (Martin, vol. 2, p. 604, Fig. 281) and the
fixed in
angles were measured with a stativgoniometer (Martin, 1928, vol. 2,
p. 593, Fig. 272). There are chances of error associated with the
latter instrument, since three readings have to be made: (1) on the
upper arm, (2) on the lower arm, (3) on the graduated arc which
finally gives the number of degrees.
We have, however, to note two significant errors; namely, SC

(interorbital width),and EH
(depth of palate). The A/P A value
for the former is 3.74 and for the latter 3.83. Let us repeat our
experiment in remeasurement in the endeavor to find why EH
and SC should give discrepancies greater than those for other traits.
Possibly an improvement in accuracy in measuring depth of the

palate may be effected by using an instrument supplied by P. Her-
mann of Zurich, in place of Pearson's uraniscometer, which was
employed in our first test.
If we again take the same 50 skulls, we find that they yield 38
measurements for EH, and the full 50 measurements for SC. The
results of remeasurement give the following data in which we have
calculated our own standard deviations instead of taking those from
other sources.
Av.
be attributed to using a new instrument. In our first test of measure-

ment and remeasurement the instrument used (Pearson's uranis-
cometer) was in the form of a letter X of two strips of metal. After
the adjustment was made the instrument was removed and the EH
distance was measured on a scale graduated in millimeters. Probably
a transferred measurement of this kind gives rise to discrepancies. In
the instrument made by P. Hermann there is a graduated central
plunger, at right angles to which are two laterally adjustable strips
of metal worked by two screws and a ratchet. After adjustment the
reading is made in situ.
With regard to SC, the A/P A value of 3.30, though lower than
in the first measurement and remeasurement of fifty skulls, is still
the highest A/P A value recorded. I believe that one difficulty
of measuring and remeasuring so accurately that the difference
between the two averages is not significant, lies in the necessity for
some estimation of the exact location of the intersection of the
frontal, lachrymal, and maxillary bones. There is often slight
damage to the thin bones forming the inner wall of the orbit, and
one has either to discard the measurement or to rely on what seems
to be a good approximation to the point of intersection of the sutures.
Perhaps at some future time an independent worker will measure
and remeasure this sample of fifty skulls. It would be of great inter-
est to know how closely his two series of measurements agree,
and how well they coincide with those I have recorded.
r-IOf-lTHOt-li-iiHiH
i-H-*t--<1<T}<O<X>00C00J<N • 00 CO
i-l <M <M 1-1
M M
CflOOS0^000005C5CiCiOi010iC5
t- 1-1
PQ CO Tf
J °°.
0) c c c c
3
<
<
H
g
3
co co co co
&
H 02 • •
BS
9
S
«
fa
CB
iJ
J
CM<N<M03<M<M(M<N1 <M CJ <M (M
fc
5 cm
oo io co io
n ns s 1 "S S
fQmfflfqfqw^Jg| "j =
Xi
5S H
5 § CO
100
Js ^ CO CO
<M <M
<M CO CO CO
<
g
3
O lO UO lO (M
&
H
«
W
K
o
a
o
fa
uo cm <m m ca
fe
JO
JO
3 3 fcfi
ESS
M WJ t>0
a
pq
^f i-4 i-i
oo oo oo oo
03 I
o >-h <M w
s 5
a
0J w
3
2 ^
a
b «
s
as
-
z
co io «o to ffl !C ffl ffl
5 CO
CO
H
J
<
s
Eh
'(U
"^
0*N lO
oo oo oo oo
Zn
03 N
§
B S -
X 1
C
S
c
p
CO CO CO CM
H
CO T*
CO i-H o
o
pa co
j o
z.
a
K
E*
as
o
a s CO CO CO CO
S
o
a
W CI Oi 05 Oi
CQ CO CO <M CM CM CO CM CO
£-3
3333333333 pq
3 ft <M
Ifja 00
2 s co
104
COOJ(MCO<NICO(M<Nl
<t
f
« :
°.
6 : g
6 2
<
g
o CO CO U5 «3
^h
co
co
co
io
co
m
co
a
z
§
H o <-H
« T-l
o
2
CO CO (M CO CO CO CO
lO CO CO CO
W 00 CO CO CO CO
^ t3 S3 S3 S3 S3 S3 S3 o
1 § § s
3£ S s m n
00
51
o.a 00 o
m en
•*#
Oi
i.O lO lO U5 lO
CO CO co co
c -^
106
co w 10 m
CO CO CO CO N N N N
— «D CO CO CO CO
co oa co co
&
<5
*
s 10
H
g
c
Z
-
J <N
« s i—i w oo
lO T-i os tj* <* co
w °
«£££££,*.*£,« g
3 3 3 3 2
5 5 S S § M pq
.0 OO
5 § oo CO CO CO CO
o
108
<M i-H y->
£N
BN
Z
§
B
r i
-a
a n H 00
o
s
o
K
O o
S s
OS
r-H CO <M CO CM CM
i-l i-H O
P3 05 0)0 0)0 o> a &* a
<o m 00
CQCOCO<MCOCOCO<MCO<MTj<COCO
10 01 c-i
J t-
ftfi W CO CN1
<M t-
o 2 r-t
CO tO 00 iM
<
a
z:
a
2 2" « OJ CO <M OJ
Z
B5
a
X
E-
o S 00
I
a
BE
-J
J
J
<
co 5 co co oo oo
E-
j pq « m
a .
> o
10 <x>
co co
U3 i« CO HO
13
C CO oj
CO "*
B co co co cO co co co co
&
Z
PS
<M <m co no
o
«
O] Oa (M CM
a w
B
§
a
OS
PS w
oq
<
a
9
a
«
3
S o
a pq pq pq
LO CO CO 00 CO <M
<
p
00 00 OO 00
£(5
O Oi
b- tH tH
s 2
<
g « g |
CO 00 OJ
§ H 10 t- .-I
o O ^J
I
I ^ ^* "^
«
w co co co co
Pi
o
a
o
« co co co w co co co co
fa
09
H CO T-t 1—1
I I
t-H CO (M i-l in co co w
g
U
t-
j pq M
oo oo oo oo
<^N us
oo oo oo oo
Zn
«\i
53
O
£ «N
w
Z
z
a
s
«
O
o
« X oo «#
fa \
© O 00 <£>
<m co co co
J
a
< O t>
fa
o O t-
Z
H
«
5
CO oo oo oo oo
<
(-*
J PQ
—
© 1-1 1-1
oo
© oo cm cm i-H
CO ~~
^CM-^CM
1-1 ^^H ©©
I
CM iJUC'^t-HO
w i-i eg CM CM CO CM in »
cm '^-'CM _|_
i-H i-H © 00 <M i-i © -* oo •<* oo ©©

s
hJ CO CO -* ©-^
O ~—'O "—
©
Ooo
o ° _1_
b 10
CO © CO © ©©©
e^> © Ol •
ifoOÔlOO
• • •
lO
03 ê>] W <M I
« 00 irt -# 10 oo
U ©
g
o
CO © CO tJ<
CO CO © ©
CO CM CNI CM
©
C\l
©
IMT _i_
«
g
K
A
!a 00
CO CO CO
£ 00 co©©
M © IO ic >o 10
CO
f- ^©©-*
-Tf ©©© ^-'CO _|_
co co 00 co CO CO CO '
lo co co © ©
© tj<
^—-© v—•©
t)<
I
rH IO ©©
T-H lH TO
© ^H .-H
3
• /-vi* ,-vt-COOO
• to <m • • •
•
•
N05M0100
— "CO w CO _L_
NffiNOlOO
^CO^CO _|_
6 w CO "*MOO
co -—'co I
co m o
6 ^ o
t*
oo
CO
as
CO
o ** o o o
co
<
a
O O HHM
6 rj -3<00
co _i_
us
a in in UO WO
3
H
BJ
'+
oo
O
2
a
o CO
OS
m oo m to co no co eg oo
co co
eg oj cm eg
woo
OO >* Oi O r-t
io .—.10
• 10 >«o
^ôo co
• •
o•
OS ÔOtJIOOOH
CO CO CO CO
O -^ CO -rf CO
.(M ^-.»0 CO 00
<M CO <M CO CO
1—1 — '1—1 <— '1—1
i-,
I
CO
P
O to
t- -^ ?D -># 5D
W
n to s—a co oo
H C oo oi
co-^co
O _j_
Z
s
o
i-H-^ IM CN1 O O
-rj<
10 ^^10 ^10 I
13
H « <N CO >0 10
(M i-H i-H
,—,00 ,~.C0 <M 10

10 •-* • • •
36 t- COOOimOOOO
JO C<1 00 CM 00
s g
4J
-a e 3! s; s: CO CO CO CO
CO -—.Tji CM CO
•00 • • •
<
im -n< <m -*jt i
-* -<*-* rl<COO.
^r- 1 ,-~.00 00 <M
ZN
no ôo-*oooo
W C0 — s
CO 'CO
< _)_
a
z
O
<N CO TfOlTfOH^
W
z
z
§
SO U5 Oi O^
E- • co • io • • •
o Oi •—-oo ~~'oo _i_
s
o
O >-i 10 oo im co o
co co co co
co w- eo"-^co j_
vOO -—sOO O <M

2 © ôo "ôo _i_ O
t- <* rji
S 3!
t-
-rj< liO .—l
o
• >-H -O • • •
z
a
s
H
si
SO»r-t • •
OS N00Nt-OO
<
0^2 CO
.»->
a co
_C3 g -^
o
.
III. AGE AND SEX IN THE COLLECTION

Fromthe total collection of 222 skulls, 28 juveniles were dis-
carded because the basal sutures were still open. The remaining
adult skulls (194) were sexed as 70 female and 124 male.
The distribution of the skulls in southern and northern groups
isshown in the table below. The age of each skull was estimated
by inspection of the endocranial sutures. The interiors were thor-
oughly cleaned, and illuminated by introducing through the fora-
men magnum a small electric-light bulb. The age periods selected
were seven in number, beginning at twenty years and proceeding by
intervals of five years, as shown in the left hand column of the table.
Determination of age was based on data published by T. Wingate

Todd and D. W. Lyon, Jr. (1924), who state that "barring excep-
tions which must receive attention in their turn, it is now clear that
there is an orderly age sequence in the progress of suture closure
and that sex, stock, cephalic index, and cranial capacity affect the
age sequence only in a minor degree or not at all."
Ages of the Skulls

Southern Group Northern Group Combined Groups
Age
Groups Females Males Totals Females Males Totals Females Males Totals
5 1 6 34 21 55 39 22 61
20-25..
27.8 2.6 10.7 65.4 24.4 39.8 55.7 17.7 31.4±3
/ 7 13 20 9 11 20 16 24 40
25-30..
\ 38.9 34.2 35.7 17.3 12.8 14.5 22.8 19.3 20.6±2
5 5 4 4 8 4 9 13
30-35.
13.2 8.9 7.7 4.6 5.8 5.7 7.3 6.7±1.
1 1 2 2 9 11 3 10 13
35-40
5.5 2.6 3.6 3.8 10.5 8.0 4.3 8.1 6.7±1
2 10 12 10 10 2 20 22
40-45.
11.1 26.3 21.4 11.6 7.2 2^8 16.1 11.3±2
2 2 10 10
45-50.
5. 3 3.6 9.3 5.8 8.6 5.1±1.
3 6 9 23 26 29 35
50+ 16.7 15.8 16.1 5.8 26.7 18.8 8.5 23.4 18.0±2.
Totals . . . 18 38 56 86 138 70 124 194
Percentages are given in bold-faced type.
The population is admittedly small, yet certain definite trends

in the expectancy of life are observed, and these trends are similar
for males and for females in both geographical groups.
—
Southern Group. In the southern group a high mortality of
males and females occurs in the age period 20-30 years. If we con-
sider the sexes together, about 46.4 per cent of the population suc-
cumbs in this period.
122
Age and Sex in the Collection 123
A noticeable decline in mortality takes place in the age period

30-40 years, for at this time the death rate is only 12.5 per cent.
An unmistakable rise is seen in the mortality figures for the period
40-45 years, but survivors of this period have a fair expectancy of
reaching 50 years of age, since the mortality rate in the period 45-50
years is low. About 16.7 per cent of the females and 15.8 per cent
of the males survive the 50-year mark.
—
Northern Group. If we now consider the larger northern group
of 52 females and 86 males the death rate is again seen to be high
in the period 20-30 years. The mortality of females in the period
20-25 years is exceptionally high; namely, 65.4 per cent. For young
females of the 25-30 year group expectancy of life is improved, yet
the death rate is still high. Males and females of the 20-30 year
period, considered together, have a mortality of 54.3 per cent.
That is to say, rather more than half of our sample population of
138 individuals failed to survive the age of 30 years.
Then comes the better expectation of life, which was previously
noted for the southern group. In the northern group, in the 30-40
age period, the death rate is only 13.8 per cent, and the expectancy
of life is further improved in the period 40-50 years. In fact, those
who survive the critical 20-30 year period have a fair expectancy
of attaining the age of 50 years. The survivors of the 50-year period
are almost the same in number for the southern and northern groups.
In the southern group, 16.1 per cent (males and females) attain or
surpass 50 years of age, and in the northern group the survivors of
that age number 18.8 per cent.
Combined Northern and Southern Groups. Inspection of the —

columns "Combined Groups" emphasizes the facts adduced for
the northern and southern groups. The totals in the extreme right-
hand column summarize the facts for our total New Guinea sample
of 194 persons.
The general trend is unmistakable; firstly, the high mortality

of 31.4 per cent, nearly one-third of the population, is in the age
group 20-25 years. In the following quinquennium (25-30 years)
the mortality falls somewhat but is still high (20.6 per cent). Again
we have the conclusion that rather more than half the population
fails to survive the age of 30 years.
But the evidence is clear that after the age of 30 has been passed
the expectancy of life is greatly improved. Abruptly the mortality
for the period 30-35 years drops to 6.7 per cent and remains at ex-
actly that figure in the age period 35-40. There is then a slight
rise in mortality in the 40-45 period, but 18 per cent of the popula-
tion reach 50 years.
The ethnological facts, though by no means adequately studied,
can provide a reasonable explanation of the incidence of mortality.
For women the child-bearing age would be from 15 to 30 years,
and in this period the most strenuous work would be performed.
From 20 to 30 years of age all males would probably be engaged in
warfare and head-hunting, both dangerous pursuits in which the
mortality is likely to be high. But at the age of 30 the strenuous
period of exposure to danger is passed. A woman of 30 years would
be regarded as an older spouse in a polygynous family in which hard
work would be relegated to the younger wives. And although no
definite information is available, it is probable from what is known
of primitive people that the active sexual and child-bearing life
declines quickly after the age of 30 has been attained. Possibly
males who have passed 30 years of age graduate into the governing
class of village elders who are no longer exposed to the hazards of
warfare and head-hunting.
Moreover, in the early age period of 20-25 years a selective pro-
cess would be at work eliminating those who were unable to stand
the strain of life. The strong survivors who had proved their resist-
ance to disease, and had in addition been fortunate enough to escape
the hazards of childbirth in case of women, and of warfare in case
of men, could anticipate that after attaining the critical age of 30
years their chance of survival to the age of 45-50 years would be
fairly good when judged by the standards of longevity prevailing
in their particular environment.
The demographical questions raised by this brief survey of New

Guinea skulls are of practical importance to present-day administra-
tion, but the taking of census returns among backward peoples is
fraught with many hazards of sampling and statistics. A student
wishing to pursue the matter further will find an enormous literature,
the statistical elements of which should be very critically surveyed.
Hambly (1937, pp. 690-696) has given a brief survey of literature
relating to morbidity and mortality among Negroes of Africa,and
his research among the Ovimbundu (1938, pp. 65-71) shows a high
birth rate and a high mortality. The data collated by Hambly
(1926, pp. 22-27) were culled from Government reports, and
British
from the pages of Saunders (1922). Pitt-Rivers (1927) has dealt ex-
tensively with the decline of primitive people as a result of European
Age and Sex in the Collection 125
contacts. There is undoubtedly a social and psychological aspect of

the problem; we are not dealing entirely with biological causes of
decline. Technical papers on expectation of life in modern society
are to be found in "Biometrika," "Human Biology," and the
"American Journal of Physical Anthropology."
A student who wants quick access to an enormous amount of
data will take as his vade mecum "Primitive Society and Its Vital
Statistics" (Krzywicki, 1935).
IV. GROUPING OF THE DATA
The table (pp. 214-218) gives the average measurements for a
number of small local groups in northern and southern New Guinea,
and a question arises concerning the possibility of combining the
local groups into a northern and a southern population. Further, it
may be permissible to consider a total male and a total female popu-
lation, which will include both northern and southern groups.
Skulls comprising the northern group were collected from twelve

localities indicatedon the map (Frontispiece), and on the grounds of
geographical propinquity and easy coastal communication some
homogeneity of population might be assumed. The factors selected
for testing this homogeneity were head length, cranial index, bizygo-
matic width, the total sagittal arc, and the facial height. The
results obtained by comparing the "t" values with Fisher's table
of probabilities indicated that the averages for the local groups
were such as might be expected in drawing small samples from a
common population. Local differences undoubtedly exist, but not
to a degree which forbids compounding of the figures.
The expression used in "Student's t-test" as explained by Fisher
(1934, p. 122) reads as follows:
x* -t
yfem — n2
A is the difference between the two averages for the traits we are
comparing; a is the standard deviation of one of the traits; and n :
and n are the respective numbers of terms in the two series.

2
According to results given by the "t-test" we may group the

figures for northern localities, and it is permissible to group the fig-
ures for southern localities, provided the data for the sexes are kept
separate.
Let us go a step further to find whether the data for northern
and southern groups may be pooled, still keeping the sex data
separate.
Inspection of averages for northern males and for southern males,
also for northern females and southern females, suggests that the
averages are so close as to warrant a pooling of the figures so as to
give averages for our total population of 124 males and 70 females,
respectively.
This first impression of resemblance between the northern and
southern populations is further tested by use of a coefficient of
126
Grouping of the Data 127
racial likeness (CRL) applied to 29 traits for the northern and

southern males and females respectively. The traits selected were
cranial capacity, L, B, B', H', LB, S, U, BQ', J, G'H, GL, NHR,
NB, O.R, 2 R, Glf G/, fml, fmb, 100 B/L, 100 H'/L, Ocl, 100
G'H/GB, 100 NB/NHR, 100 0,/U, 100 G 2 /G/, 100 fmb/fml, and
the B angle.
The formula used was as follows:
Let the averages for the traits of the two series be a, b, c . . . x, and a', b',
c' . . . x'.
The numbers on which the averages are based are A, B, C . . . X, and A', B',
C . . . X', respectively.
The standard deviations used are o-a, <rb, o-c . . . <rx, taken from an Egyptian
series (Pearson and Davin, 1924, p. 338).
Then
AX A'
-(^r X A + A'
This alpha value is worked for each of the 29 traits under comparison,
and the CRL equals the sum of all the alphas divided by 29, minus 1.
The coefficient of racial likeness for males of northern and south-
ern New Guinea is 3.60 ±0.17, for northern and southern
and that
females is 2.90 ± 0.17. The probable error is given by the expression
0.674489
v/mO-h)
where M, the number of traits, is 29.
In order to compare the samples so as to give the coefficients

the value they would probably have if each group contained 100
skulls we may use the formula
rii + n.
Reduced CRL= crude CRLX50 ni X n>
In this equation 50 represents , and ^ and n, stand for

100+100
the average number of skulls on which the means of the first and
second series are based, respectively.
So, in order to reduce (adjust) the coefficient of 3.6 we have the
expression
81.31 + 33.00
3.6X50 81>31X 33.00
which gives 7.67 as the reduced coefficient for male groups, and by
the same method we get 11.84 as the reduced coefficient of racial
likeness for northern and southern female groups (Pearson, 1926,
pp. 105-117; Morant, 1931, p. 214; Wagner, 1937, p. 120).
By number of skulls we allow for the fact that,
using the average
owing to defects, not all traits yield the same number of measure-
ments. For example, in the male population there are 86 northerners
and 38 southerners, but the average numbers of measurements are
81.31 and 33. For females the average numbers of measurements
are 46.31 and 16.65, for northern and southern groups respectively.
Coefficients are affected as shown below:
(1) The reduced CRL (male) is 6.83 when full numbers are used,
and 7.67 when we use the average number of skulls measured.
(2) In the same way the reduced coefficients for female groups
are 10.84 and 11.84 according to use of the numbers of
skulls in the groups, or the use of the average number
of skulls measured.
I have not calculated the actual standard deviations for these
averages, since my series are long enough to assume that the stand-
ard deviations worked out on a long Egyptian series of skulls can
be used (Pearson, 1926, p. 108). Stoessiger and Morant (1932) also
give a long list of standard deviations worked out from 199 old
English crania.
A student who is not satisfied with my use of standard deviations
from the Egyptian can work out an approximation. The table
series
(pp. 214-218) gives averages for 18 groups of male skulls; these are
g
partial averages centering about the general average -, where S is
n
the sum of the terms, and n is 124 (the number of male skulls). The
method is explained by Udny Yule (1937, p. 435), and the formula
given is:
Estimated standard deviation squared= —
n-1
T S (x— x) 2
Take the width of the nose as an example. By finding the sum

of 124 measurements and dividing by 124 the average works out at
25.3 mm. Now subtract each of the 18 group means from the general
mean of 25.3 mm., then square each of the differences and find the
sum of the squares.
The sum of the squares, that is, S (x — x) =11.22

2
The value of -i- =

n-1
—
17
Therefore, by substitution, — 22 =0.66 =
11
: the standard deviation squared
Therefore a= V 0.66= 0.812
Probable error of a= n=X

V2n
0.674489=0.09
Probable error of the mean is -?= X . 674489= ±0 . 13

Vn
Grouping of the Data 129
We have, therefore, for comparison
(1) Pearson's standard deviation for nasal width (893 male

Egyptian skulls) =1 . 77 ±0 03
.
(2) Our own standard deviation for nasal width (124 male
skulls) =0.81±0.09
(3) Pearson's probable error of the average = ±0 .04
(4) Our own probable error of the average = ±0 13 .
Naturally the probable error of our New Guinea average is

larger than Pearson's error, since Pearson measured about eight
times as many skulls as we have in our series. I believe, however,
that his standard deviations will serve our purpose in subsequent
chapters.
Q
H
O
"Z J o o
<
<
w co co
O
o CNI
Oi
p3
2 s
a <
Q X
z
x
GO 04
lO 00
CM CM
« + CM CM + to W lO U5
< =5 I I I
I
b "b b b b b b o o o o
e a)
3 3 3 3 3
pq pa pa pa pa m pa pa pa pa
CO
C.Q CM
3E ^ co
C3 t3<
c
130
Z 10
Z -*
J
«j
j5
::::"*.:
. . . . 10 •
°°.
oo
: : : : :
®
co
N °*. x
«dj
Jo
u
:
. ...
: : :
MS
CO
. :
•
00
00
: : : : : .
CO
CO
N O
,
j?
°
....-co
•
•
•
• *tf
CO
•
Ci
-co CD
co
«::::^:^:::::^
O • • -co • <* CO
-iOLOoomcomo<Mcoocot-i-i
Q
a
o
z
<
«
S 03
S
W
0H
13
O
«
o tf.... «. _.... t C o
< 9§••••05-05
•© • • t- • oo
00
o J
. . . .
f_ • Oi • • • • -co
^
O. • •
.
•
•
-co
-OS -oo
• io 00
oo
7]
s
Z Tf
j-t 05
t- t-
CO CO OS
o
O <M
<T
H
g
o
+ +
z
«
a
S
£ r
b r
b r
b b
r r
b r
b'bo<-oo
F-
•3 bi
3
pq
—
ca
J3
I
o
& an m
2 2 2
S o
+*
o
*»
< jS
5 ~ — —
ca
C8 ca
m ,-b .-a
O U Tf 00
CO o to
C3 r-,
to m o
to
W oo
a
S
H
£ •-
5i
r^ >>
O t-H
+ + + + +
oo f
b r
b p
b r
booo r
b b
r r
b b
f
a s a a a a a
"Bo So d c d
o pq p
l-J s § § s § §
S =
•SB
3 §
^
to
^* ^t* "~f **3*
136
Oh
o
«
o
-1
CO CO CO CO
3
H
Pi
3
<
a
CO
JO
Jo
a
H
O
<
as
<
Q
H
O
z
<
OS lO us >o
< H
CO
HcacoeO'^'wsaoeaeN'-t'-Hi-iso
2£
?!— CO CO • tH © © OS GO CM © CM
o 6 lO
CO lO lO lO U3 tC
H J
% >*
a i
«
a
<
g
3
a
is
+ * + + + +
a
Z
z
H
n
Soo f
b r
b b
r r
bo o *b b "b b
9
S S S S S
So & d
i PQ
§ S § S S ^ £3 ,& JD ,0
3 fe T-l
139
co co co c>i
Oh
P
o
«
o
<
o
h-1
eo co rH
lO US CO 1Q
CQ
i2
E 02 co io
§
S
»H © 1-H
Eh
a
o
w
CD 00
<
e I I
5
a
+ + + + +
H
55
i S o o b b b b o o b b b b
H
S
3j C C G C t s
i- >—
S
o
5
cq
3 3
O O .g
-n -t> .TS
^- — a
e« <*•
0,
p
o
«
O
J
<1
o
o
Q
W
cs « 22
<
<
W =5
P3 go Oi C5 Oi d
"
3 M
O
w
 ?0
o
Q
H
C
pi
<
S H
CO co
SI
JO
S«
Jo
J
6
«
6 «* CO CO CO
o
O
<
o
(J
iH O H
£ ^
co co co
10
co
co 10 cq
H 2
w
2 *
1 £
w
a
W
<
c + +
is
a
is
K
§ oo'b r
b f
b ,,
b r
b p
b p
b F
b b
r ,
bo
E-
BJ
C
r3 £ £
2 s 3 3 3 3 3
9 ,Q ,0 § § s s s s s
•3§
^4 Tj* ^* ^
145
o S
T3
W O
2:n
0.
3
O
BS
O
j
<;
o
s
3
=1
3
w cm
CO CO CO Ol
W -3
C3
u 0) in in co >*
I I
z
M
a
S
H
oo p
b r
b f
b p
b boo
l
QC
o
S 3 3 3
o o
n w m m « «
in
m m in m
CO
148
IQ IQ *G tO
(M <M CO 00
lO Irt lO to
J o
B ™
CM 00 CM lO
ca 3 I.O CT5 i—
& tr
oo oo oo oo
m m in m
5 a w
CO CO CO CO CO CO CO CO
K 9! A O Qi
p
Jo
p
o
c
o
-J
cocococococococo
CO <M <M <M
- w
O -<t rH
i— I 00 tJ<
o
#
O
<
o
o
<
5
3 a
'§ <
8 I
o
W O
I O OJ
2^
w
g
<
2
s
o
I I
a
E
Ofl> r
b b
r
b
r
b
f l<
bot-ooo
H
o
>>
£ jbtf
J*
3 3 3
g
,3 § S 3 www o o
CQ m M
5 § co
151
o 2
o t-
p
o
O
«!
O
o
Q
H
o
2;
<
<
CO
H
a
<
w
H
a
S
a fa 2
a c^ci
I I i i
a> >0 ia io co m
<1
I I
r
b r
b f
b r
b booo
r
>•
3
jM
°E
£
'E
£
'E
8 3 3 3
,3 S 3 S pq pq pq pq pq pq
CD
M<ja o
o
CO CO CO CO
152
o
as
O
Q
a
e>
<
es
PS
<:
•<~
M N H H
&
o
B5
O
<
o
Q
a
« ©
a co
co i-i t-h <r>
t-coa>LO00(MP0CDi—it-i—
fqCMcooacoesicococococNico i
H H
J t-
K
o
i-H t-H «D
a
g
C
is
+ + +
a «! 5 w co co in co <m
«
a
H
S'b r
b r
b r
b r
b'b r
b r
b bc>ooooo
f
<JO<J<H<J<i1<J<1<3 <J <J <J <j <j
2
OS
§
a
co
^ cocococococococo
154
4< -rt<
C"j C"s CI CI
CO ?D tD «3
S i-s CO
Ci 55 C *:
OJ CO <M <M
« OQ us \a
cocococococococo
a
W OS Ci Oi Oi Ci
u>
Jo
P
o
«
<
U
C
U5 (M i-l
S
OS O i-i
CO CO CO CO
H CO <M
Is
<M <M (M <M
x
o
<
H
X
P
+ + + co m co m
&
Z
z
r
b r
b l<
b F
b r
b b'b
f r
b ,
b o o o o
«
H
SS
O
Z
3 p
** t
<** < t t
<£ t ^<<<t t <<<<<<
cJ3 O
3 6
156
o
©
-,
35
Ok
13
O
a
O
<
o c ^Q
o CMCMCMtMCMCMCMCM
a
u
z
<c
3
<
s
Slo
CD CO
1-H O
O co
o •
o CM
a CT5
6
u ^
00
lO 0> CO lO
00 0O 00 00
a
55 „
tf 00 us co co co
+ + +
« r
b r
b b
, f
b r
b F
b b
r p
b r
bo ,
'
<> < >oo
<!<J<J<;<i<!<J<t1<3<5<;<iJ<J<3
o
Q
H
O
S> to
J a
I s
8 H
fcN ^j.
O
I
co
H
£^1 <X>
to
m
£ a
to ©
Oh
P
o
o
3
o CI c-
a
W
o
<
erf
j] N »
cs
lO «s CO
>«*
< to pgC^CMOJC^CMCNICOCOCNlT-HCNlCNieOCOCO
<^D CjO
5
o
is M CM + +
H
z
a
w S o o o *o b b b b O O O ^ *q
n
H
2 o
«sj <t) <J <j

J CO
o 2
p
o
«
O
j
<
o
iH CO W
P '-s <M
« 22
C oo
03 <M
tn
v.
a
H r
« &
P
w &H"
H
rl
M
h
T-H CO LO
+ +
b b b b b o o o b b
3 3
= <c < muiwwtnmwui W ^ ^
i—
I 00 m
;£ CO
%* s
161
36
as
W CD
6 ^ co co co co co co
<M 00 01 CO
r-H 00 T-H 00
co co co co
CO CM CO CO co co co co co co co co
w
E-
Z
H
s
S «
§2 CO CO CO <M CO CO CO CO
+ +
I I I I
b b b b "b o cm- o "b b
<J «J
£ B
I
00
P
o «D t- t- C- t-
j
o
o <M <M IM <M
-1
H
U £ w
55
<
H oo
O a,
X
g
O H ,-1
g
3
a
+ +
I I
tf
H
E
oo'b f
b b
r r
b boo<-
l
o *b 'b
H
< < < <

minmmuixnmmw
E
C3
C3
O
85
2n •
00 OO 00 00
P-c\J
oo oo oo oo
Efi
Zn ©
B
B £ wn ©
w
5 B
<N ^
2 ~
2 I
©
t-
X
a
a,
M
"3
CN1
>"
CJ + +
B I I
3
c
a
o b
l r
b b
r r
b F
DOOoo b
r p
b
J?
« o
a
s
< < < <
c
£
5
o
as
a
a
<
t~50t-O00l0->tfC0cNlC0C0tf3-^<OcM
caa>oocooiooooocO'-i«Dcoo-rfit~(M
t-ooooeo<M-*wc<iy3oot-oooo
M<M(MO0COO0C<3<MCOC<ICOO3CO(NCOeO
W
H
w
< «?
E-
w
o
<
+ + + +
H
Si
w F
b b
r r
b b
r
b"b
p
b
r r
b r
b b
r ,
b r
b F
b' b
<
r
b
E-
8
3 3 3
a
as
'•3 ="3 cij a!
^££££££££:££:£ as
& * *
166
J o
Q
a
o
z
<
05
^ <
S l
I
o
&
H
i
>-
CO CO CO CO
Q
H co co co co
O
z
<
<
s
fo
T
w o
z *
00 i-H i-H
02
6 2 rj
<M<M<MCN1<M<MCN1IM
+ + +
<J O
I'b'b'b'b'b'b'b'b'b'b'b'b'b'b'b
3
Q
H
C3 CO CO CO CO
z
<
X
<
^H i-l O
'§ <
V J
GO o
Is
><
E->
K
O
w
< W CM
a
z
O
+ + + +
a
2:
z
«
a b b b b b b b b
K *b b b b b
H
«
O
3
U t"
<M OJ <M CM
o S
CO OO CO 00
o
* oo
o •
<
o a 00
o
5 OS
6 oo
t- lO
oo oo oo oo oo oo
io ic m io io m trs io
DO
z »-H CO Irt
cm co
Z io
2 S
>-
H CO ID CO CO
X
g
W 00 CM 00 CM
-<
J ,-H
H I I
+ + + +
a
z.
a
i'b'b'b'b'b'b'b'bt'b'b'b'b'b'b
«
3
-
£ £ £
£
«* ^ i-i
CMOCMCMiOlOuOO-'tfOO
CMCMCMCM'-ii—ic<ICOeO-#
w
a.
-'
o
«
<
o
o
JO
Jo
o o
Q
a
co co co co
<
pi
CO CM
O i-H OS i-H r-4
a
<
6 t2
K
<M(M(M(MCO<M<NOJ
Eh
K 3!
t- t- t-
Oh
o
o
£ »
Q
H 2 rn' J,
o
<
32
si
< X
O i-H O r-l
w O
DQ
fe
ON .
oo
an Zn oo
co co co co
£ H
so
a 1
00 oo oo oo oo
H
ac
< + + + + + +
a
2
S
u
>
a
bbbboooo o o "b b b
z
« >1
a
S
o
a
a
o
<
PS
«
c« , o o
Z S3 ® "^
H
§
SO CO (M CO
la
2 8 » <M T-t Csl i-t
JJ'b'b'b'b'b'b'b
^ .. > .> „> -> -> - > „i> .> «> -> . > «> - > ->
& SS 33 SS §« §S ^2 §S 3£ 32 SS 33 3'tf 3« ss
8 u ^ ^ ^ ^ .s ^ ^ .a ^ £ ^ jM •- j«
j
.55 .ts .53 .fcj .53 .5a
£^
W '3,W 'S.^ 'S,^ 'S.^ '3*^ 'S.^ "3." '2.^ 'S,^ "ftW "p,W 'S.W 5£
.is
'S,* 'S,
.fa .53
xfiuiinxnm<nuimmmuiuiuiui 0) O) 01 cu o>
3 b
o
O
<
o
o
CO CM CO (M
Q
H
o
Z
< « 55
OS
S
<
.§ 3
'2 < <M <M <M <M <M <M <M <M
8
Z p
H 1
5 w*
05 &
H §
S3
CO CO CO CO
w
<
|
+
to
CO <M <M IM I I
I'b'b'b'b'b'b'b o o o o o o
>> «.£ :> -.£ -> -.> -> -.> -.> -> :> %£ >.> -.> ->
3 .53 w .53 ^ .53 ^j .53 v .53 v .53 v .53 k£< .b >£ .53 >^ .53 Jjj .53 m .53 v .fa jjj .53 v
k>^k>:33k>:35K>733k>:33k>r33k>:33k>r3k>:32k^r33k^
J9 ^ a* a M a w a w a* a* a M a* a^ a w a* a M a w a
IIS <£> ID <£> U3

co co co co
2§ co CO CO CO CO
179
SJ
,35
:::::::
::::::: ~
gg
o
§«:::::
JO
•«#
^
BO
-•
CO CO M
Oli
j
>
o
:::::::
. ..
•
®
-CO
• .
oo
:
• as
-co
°.
03
o oo co • o
g,
< "g t- 00 • r-l
o O • • • • • co co •>*
o
n
cocococococococococococo
s
<
s
s
1—1
O oo co
CO Ci CD 00 OS o o co
o ^ t-
oo
co
oo
oo
oo
^h
oo
ôcocoi-HCOoot-io-**-^ • •**« co
a, ^ o t- »o o -* Oi
oot-cocot~coiooococ-
uo co t~ i— i
•
Tf
oo
o
co
o
BS
- • . . • • O) N -OS •
-Oi
<:
o
o
«
• • • • • -COOS
0000 -i-l
• OS •
-co <J>
o
-;
p> . . . . • © C- • Oi -Tj<t-
^
u COOS
oooo
•
<M
-oo -coo
-oooi
-jCOCOt-OJCDCDWCMCDOOlOOi'st
QW-«3<CMOOt-iiOiOa5iOt-^J«<M©
oooooooooooooooooooooooocn
00 (M (M 00
lO lO LO U3
,-1 lO
^ lO
K
u
<r
H
z
3
<! to
+
a
z
a
= $'b p
b r
b r
b r
b r
b
r
b o o o o
E-"
!
2 > „> „> -> „> .> «> »>
'
3 3S 3^3 3a3 3 '^3 3a3 3a3 3£
w.S
I— ^L-t-I-.i-t-HS-MS—
w.2 v.3 w.S v.sj^.S v.Sw.S — l—l— w.2 w
v.S v.Sjsj.Si/.S ;-<
3 H aw a — aM a ^ aw a 1 1 5*' aw cw o. w a w a w a* & w o.
II s CO CO CO CO CO
si s co co co co co
182
C
CM *-H ,-H
- t-
o O
co
to
2
o
o
-J
Q
H
O C2 OS "3
Z OS OS OS OS
<
OS
<
2> w > OS CM
m
« oo os
z
ca CM
1-1 CO T-l
-i (/J
> kL
3 £
Z SB
H £
60 CM CM +
< I I I I I I
r
b r
b r
b F
b b
r r
bo<-o r
b r
b'b r
b r
b
3 3
o o o 3J c
cu£ g
'S, 'ft 'ft 'a <! -< <j <J
0) cu 13
QQ
X CO 02 02 en
0)
"I s m io io m
5 § co CO CO
""^
CO
"^
CO
^*
"*vj*
184
CO ^H i-H
id i-t i-<
-
«
j a •
< O CD CD CO CD CO co in cd «o
o
Q
H
O
C~. C. C OS Oi Oi OS C5 <J5
t3 -
DS C71 CJ1
CQ CM <M<MCOCO<M<N1<MC<J<MC<1 CJ>
x j CD
Z
a
1
os
S 03 "^
cocococoeoeoeoco
»
g
g
3
+
2
03
f
b f
b b
r f
b r
b boo
p r
b
r
b'b r
b r
b
T-
2 2 -T.5 o
3
o
3
O
3
O
r3
§ §
'a '5.
'S. 'a, 'S. 'S
CQ CQ CQ CO CQ CQ CQ
Q
>*>
°
JO
cu
o
o
-J
<
5
Q
H co co co co co co co co co co co co
a
<
Pi
s
<
S to
CO lO U5 CM CO <M
co co co co co co co co co co
OB
Z
9g
2 i
C0<M<N)<M<MO3<MOJ<MCSJ<M<MOJW
HO
a)
<!
+
I I I I
«° f
'b
r
b r
b ,
b b'bot-o
r f
b r
b p
b f
b F
b
c c c
o £ c e« c<5 S3
§ s § § JHF§c3
<£ 3 p
up
W CO
n. 'S. 'S.
<J <3 < <
CO t/2 TO CO CO
II s CO co co co co
31C I? co co co co
3 ^
186
J oo
a,
CO CM OJ CO
U "3
o
-
5 x
=
o
H
H
Z
H
2
S5
a3
H
X
E-
05
O
z
Ph" 00 t-
I 1
do o oo oo oo oo
BE
H T
£N
CO "
S <NJ ° co co co so
u
Q
a
O
z «
a! J
<
a
<
is
H" W
H
H
K
O
H
(M CO 00 CO
> ^
2 2 j g
ai
>
E-
£
Eh 5 S r-J
<
a
3
o + + +
a
fc
55
OS
a 1 "b "b "b "b
r
b r
b r
b b
r r
b r
b r
b b"b
r
9
P
o
•S « tf r3 C3 Dj
a
s 3
£
a SSSSSSSSS
3 b
60*
5 t^
w io m w io io io m
190
th ia ©
CO t-
JO »-<
"OS
3o
i-i Ol 00 i-l O • • • • • • • -<M
m f N O
Ô0505i-H0
^1-^t • • CM
o
J Oi
CO 00
O
O 00 IO
L—
CM
t-
i-H in i-i
p
o
Si
O
50 CM IO
<
IO CO CO CO
—g
"<*
o CM CM CM CM CM
e rH «3 O O i-i
a E o co co o
o "~
i-H
co co co co co
<
s
•33
13
to
s a
© a
00
M
a
a
S
a
g W
p „
w to O O
< H
a a
5 <
> a
CM CM CO CM
+ + +
CM CM CO CM
"d 'b 'd -b

r
b r
b"b'b f
b r
b'b b'b
,
>>
CO i-l i-l
t^j i-h lo
p
Q
O
J
g
o
Q
H
O
s §
§ rH
CO
—
G3
—
£
a
a
o
•z
<
pq m pq PQ ffl
<
s w
•I <
§ I
? ^
CO
z 1-H O
H
H
oo oo oo oo
H
1 O
W
ZN
<NI g
H
gjj
r.
Eh
<T
H
Z
o + + +
W
«
a
J'b'b'b'ba-'bib'b'b'b'b'b'b'b
as
H
O
03
« tf 3 3
c4
§ s S S S S S
pq pq pq pq pq
O.C t— fc-
CO
m m m lo
195
p
o
as
O
j
<
a
o
o
H
O
S 2
00 <M
3 9 M
..
B CO
J 3
H
1
*
~j 3i 35 ~.
I §
p _
eg no
r w 1
>-* t-
hh 03
p >>
- -3 00
Eh
—c
< O
a
2
3
o +
a
Z
z
SS
a b'b
r r
b f
b r
b p
b b
l f
b r
b
n
f-
P
O
:/:
5 Q>** a
t< 3 v ? c3 c3
I -3 3 3 .2 S*^
m
o3i2
-
3 2
nifri QftQft
a >^ >,^
|
> g s _ s
cob
.-3
_.
s § s § § m
+J +» â w
tc i
£ 3 CO CO CO CO CO
196
X ^
12 lO
0.
5
o
a
<
co in co co
o
a> Oi os co
05
•5 < P5
COCMCOCNIC<1CO<MCO
IO IK
ta
Z
H
2:
w
a H
a
a o
5
U3 UO UO io
> H co co co co co eo
; h
g
i
S
E-t Ci CJ OS Oi
<
g
+
« a> O O o O O 'o'b'b'b'D'b'b'b^
w
X
£>
o rt c«
72 X3 « h
c
ca
c -.
<^ o<
rt
^ >^ >m > £
•~ £ .2 5 K 3 3 3
c«"cp5-c«s
erf erf
s § a CO 72 72
So2o
3>s S si
CO CO CO CO
5 § CO CO
197
JO co co
05
S
JO
Q UO
B i-t ^H
O co co co co
55
<:
S
<3
s §
•5 <
s a
© a
z
6 N co co co co
H a
03
a
a
1
a w uo CO
3 H
W 03
s as
<M <M <M <M
1-1
£
5
H
<
a
z
5
o +
a
z
S o o o f
b b
r r
b r
b b
r p
b f
b
r
o b
r
a
K
E-
O
co s <D
"3 ?3
"E °"E °
§ s s O O So 2o
CO CO 3 3
PL, PL,
a a *°
5 § co
U
198
m co w co
Pi
C
<
o
Q
a
o
OS
172
CO
O El
H a
H
P.
a
en en
> a
Eh
g
5 in to co <m
<
a
-2
+
a
Z
as
a
x
oooo p
b r
b r
b p
b r
b b'b
r r
b r
b
H
3
EC >>
IIS
O «>
o £
O
^ 2 O
Oh o
O
«: o
° 2
o 2
Q
a
o
z
PS
<! a u
"§ <§
.§ 3
o B
? fo
r.
z
-
z
a E-
r DP
S
a a
D _
T3
o £^
c3 ^ gj rt
c 2 a+J
+j-fl
cu'C
^! ai tj M
Pot to 'G
O M 2 Q)
co 2 B +->
c-« O) S3
2 £ b.2
"Si St 8 5.-
pq PQ<l5 opq
s
•S
•J
«!
oo oo
o H
fa
T 6
cs
S3 w
£N
3 H
P3N
sc
u
S
>
H
3
o +
K
a
oo r
b b
r
b
r r
b r
b b'b
r p
b r
b
of
& T3 t3 T3
<tf
mo
a, """ *"
c3 "a3 ' *a3
rt
S3 ,So J2 cy ,-3 a> *-
\1 2 S3
»-< u-5 wT a ,rt,.i3 .S3 .S3 .S3
»s 8"G g
§ § § 00 02 CO S3 3
Oh Ph
OS B
201
O O -^LOICOIOIOIO
CQt-i,—ii-|i-Ii-iCNCMOi~IC4
„ t—
. OCvltM'-HOOCOIMCO'-iOlMOOi-H-rf
gc^r-H(M'-HOOcai>a<MCj--Hi-iOi-H
c
P
o
o
c
IOOÔÔC^^CMCOIOOt-IIMi— "3
WCMCMCNJ'i-iCMCMCMCMCMiOCNlCOCMCM •
,-HO"*CMSO«OC<3CT»CJiO>
jOOOC7JO>CTiOC7}CJ}C75
m
^cNloo->*ooa50locoo-^
S-O3<MC0CMCNJCOCO<MCO-<*
,
•5<
© w
fa
f
WoOOOGiOOOlOlOOClOOOOOSCS
r<
H
g
P
w
..
tow
H 3
MiMCOCMCMCOCOCOCOlMCOCMCOCM
ro «* CM CM -sji «r> ee LO «5 cc
E? K
H iJ co
I
><
i-i i-H as
H
3
a + +
ft
H
i
«
B "b "b "b » » r
b f
b r
b r
b f
b r
boo
P ctf rt c3
o i^^nii<M^<M. 1 CM .=. CM r=< CM .=- CXI
i P £Q £Q g-Q §<Q £Q §P g-P £P gO £P §P g-P g-P g-

E"C B'S 2' c B'C S'S B'S E'S 2'S £"£ 2"g E'g E'C E"g 2
3 3 3 3 3 3 3 3 3 3 3 3 3
pLipLlfapHp-ifafafafaPlP-lfafa
SB £
202
a *n
kO 1-1 rH
J 2
ft,
P o 2
o
«
O
j
-
<
II
© a
U &H
I
a
g i P o
s I ira io io io
a u
a 3
g
P
W
W „ P oo
<
BO
H a o
a -
> a ^
E-
5
Eh
w en a oi ci
<T
a
z.
5
a + +
a
Z
z
pj
a
=
,«j b b *b o o p
b b
l F
b r
b p
b boco
l
a 0]
o I
>» o o
< h
H 3
« N N N CJ OJ 04 <M
a
Eh
<
H
Z
3
a _l_ ira
CO
co
I
co
I
_i_
z
as
a
3
^ 'b 'b 'b oo p
b r
b b
l F
b'b F
boo
71 71 71 71 71 71
o
71,
+J
71.
-l->
7i
•*-» ,
71.
-U
7i
-U
71.
-U ,
71
-tJ _, +^>
7t
__ -U _, -^ ., -U _- -U „ +J _, -t> -,
I Q gO gO gO gO gO gO gO gO gO gO gO gO gO g-
g "C g'E g'E g'E g'E g'E g'E g'E g'E g'E g'E g'E g'E g'E g
^ So So So So So So So So So So So So So So
Pl,Ph^hPh:PhPh:PhP-4PmPhP-iPiPhP-i
OC
«g £
204
CO N W IN <X> CO
CM IA
05 Oi
O i-H i-H 10 oo
co co
co oo o CO lA CM
£ oo r-i
CO
00
CM
t-
CM
t-
CM
tJ<
CM
CO
CM
Q
o
z
•<*
IS
«
s
1 <
•5
a 1
5 w
Eh
Y
S H
i 3
OQ
< a
BQ
a _i t-h i-h tr-
< E-
3 2 O m
> H ia
3 >>
£
Eh
<M CM CM CM
a
z
3
O + +
< o
Z
«
a
H
^ b
r
b'bc><-oo
r r
b b
r F
b r
b r
b r
boo
f*
CS S3 03 03 03 c3 c$ d c3
o
I
I
-^ , -*-j
^ -^ gJ -*^ r.j .^ C j
Oj
|C<l^<M~<M^(Mi?CM
0> cu
i Q g-Q
<!j
&Qop op op ap op ft Q op ap a Q op o.
q.q
cu 09 <1> Oj
- -
o "S B'S 2"g B'fi B'S B'S Bg B"g 2 8 B*gj B'g B"g 2
^ go So Sog*SSo So So So So So gSo £So So So So
PmPhPhPhP^PliCIhPLhPiPhP^PhP-iPh
N x* OS T* 00 CO
00
N
.-I
i-H
i-H
CD 01
IA IA >A >A lA IA CO CO
CO CO CO CO
205
at
a
2 ° co
lO t-i <o
CXI •<* t-
Ol C5 C~
<M C<1 <£> "3
J
•«:
o
o
00 CD 00 (M
Q
B
O
2
<
<
S <
« i
x a
fr-
a
a H
iz
OJ 00
s
a a
bs H
in
P .
W CO
3 a
a j
u
O CJ
m
CO H CO r-(
<
a I
z
3
O + +
&
a
a
n
o
03
206
3 w
g w
03 03
H 3
J-< O T-l
CO CO CO CO
ft,
D
O
o ^ ^ *
<
o
o CO CO «o
1-5
72 CN O i—I CM »-i CO CO
^ o
H BcooqcocM'—icococococococmcoco
so 4D CO
• t~ OS 1-4 OJ CO
CO
Z
ta
as OS Oi OS
Z w
ffl CTJ
H
s
H
p . ^ OS CO CO CM t—
< H CO CM CO CO CO CM
9 °
> H
Q >*
Z H
£
3 o O
01
< o
H
z
3
o + + +
z
04
W
K
r
D r
b b
r r
b b'b
r
b»o
r
E-
P
O
c3c3P3c30303ctic3o3c3c3q3CtiW
^CM^c<i^cNi^c<i^cNii2coi^co^co^co^co^co^coiicO'-Hco
72
| Q §<Q g-Q g-Q §<Q §<Q g«Q g-Q §<Q gQ gQ g-Q gQ gQ g
5 -c g-c g-c g-c g'fl g'g g'g g-c g'g g'g g'g g'g g'g g'g g
J So So So So So So So So So So So So So So
^ s p ^ 3 ? 3 3 P P P P P P
=3 E (vi
208
H, © iH
o oo
t3
ST °o
<M CO CO <M
o H
? fo
w Z
a D O t-H CO OS
a
H
X
o
63 j
> H
Q >«
Z H
w oo os
<
g
3
5 !M + +

f cm
U3 -*
gfa
< W
O <c
> H
2 £
Z E-
1-1
£
s
Eh
<r CM CM CM CM
H
g
o <m + + +
Si
H
z
OS
a
s
H
o
oa
211
CM IM <M CM
Oh
P so
O
O
< O •
• • • * • * o • CO W • CO lO
o
o
o. . . • lO O • -># C- •TflO
-OOi
^ 00 Oi • 00 t-
a» • • • • • • t- i—
i -co • • cm oo
JS •
Oi 00 -* OI t-i
w • •
-00
. • • • •
. . . . • t- 00 • 00 CI
^CMOicot-Oii-Hcococot-ooôco
w 2
rti—ioocMoooootoajococoocoi-i
^cjoooajoooot-ooooc-t-oooocj
H
•5 <
o «
w
x H
t- OS
§ a
H CO t-
g w
H 3
<:h
Q O
> S
5 lO 00 i—
<
W
z
3
o + + +
OS
H
K
&
O
m
212
.2
a :0s
to
g i* o
a
T3 ^
S3 g ea
rt
3 J5£
*o o +j—
,Q -^ OS +J
gj3
g 13 ox;
a;
SUM :0 ^3
ks° s fl co
.2 °"C
S bOr? +» =3
*
<n O 3 - eg.
[JC M § bfl.2 a>
o IE 2
** G p^ ;
a S3TJ £=0 a .8 * c
h rt S3 & O ci ea +3
rj- Bj
o .15 o> t3 "c
z
<
£+^ «a g
*h
g
J-
g
(h
legs :0 :0 :0
« Sill Oes tf£ £ £ £
li*
IS
© H
ta
z S
60 DO
BN
N CO t* r*
> H
H
s _
<
H
Z
3
o + + +
H
Z
« 0) O o O O 'b'b'b f
b r
b r
b b° -o
r ,
S
H
o
CO -a3<MrS<MTROJ'5SOa'Sl« -TlCO^co-r-cO'^COrr,
4) <L Qj Qj OJ
CO *- 00 -- CO •- CO
0) Qj
— CO
CL
3 P o.n ftQ ao ftQ a q an
3
ap
3
ap
3
&Q
3
ap
3
ap
3
<L
33
q.q ap o.
3
g £ g'c o-e o-c g-n o '£ o-p o c P"E P'S P'E P'E P'C P'E £
-
J 2o2o£o2o
3
Ph
3 3
Pl,
2o
3
e«A
hu
Ph PL,
=1
PL,
333333333
A
ca^ ca.t a
Ô bj± t« % es± ca rt ea rt
(-O l.O >-0 <->^D r t-O >-0
PhPhPhPlhPhPhPmPhPL,
«-.0
.3 3 CO
213
~ T)iTl<oat-NMH(10OHiO«0IMC0«)OOt-i000NTlitr-00lM00iO000J00W00O
w ••
2 o«OHoot-a>ooooa!t-!OiOTfio<rit-'*Ti<M'^t>oot>HHOooooioooioo
w
£ HH»MOOiHooHOoot-t-u5t-[-ooimoî>a>t-W(M(NHHHHfflt-fflH
pit-OOWOOiOOHMTl'Ot>ina>(0'ÔOHtDCiO'*i0010N'J OOt-CDHOOMCl ,
E oooiosorooooHoost-t-^^xct-WTjiTjitxst-irtMHHOHOiffit-tflo
hj
, NOOt-OOOt-bOHOU3OMi0®Oi0NINOOMOT|iOOHC0M"3O'* •
O t>in05iOHCON(0000«Ditt-^'*!C!Ct-t-000>ON^BH(OOOHtOOCl
aoaiooooojoooaomoiffioioiffiffiooaoooortoffloaoo
I— 1— T—IT— 1— 1—
I T-H 1— I1— T— t— H T—
I I-H I—
1-H I t-l I I I I T-l T-H 1—1
»U30NOH«OOWiOHiOiOOifliONOOt-!OHOOOtl0^fflÔt-WOt-
i-b oaioooooo(£iot~oot-t>a>coKioON<ooooiNWOO)^<wo5HOoooot-Ht-
MlNCOMINNNINNNINHrHHrHHIMiHINHINIMM'HMIMIMCOCOHiHWIMrH
p
o
ÔMtCNtOWNO^NOt-iOt-ifltÔOt-HOOOOOSOINtOiOiliHOiOtDt-
«
(fltot-tîfltowTfoi^mcoaiHt-Hoitoowi-ic<ic>]cococoiot-t-T-i<M05-'tfeo
o O5O5OJO5O5O5O5O5O5O5O50000O500O50000O5O5O5O5O5O5O5O5O5O5O500O500O5O5
J
<!
O
o
Hi0t-O»HiOT)iXÔt-MON(MOOOO05C0Nt-t-<0t-rtiTliOt-m'^'*
PS
C CDOtOHOOOrtHa^Tft-HlJitDOMOOtflttHOIMHOOHlOaMlflOt-OJM
O P5oloioiaiajooffjooiaxnt-KKOooooooowooooooQmoiasoaiajooffjoooocc
OiOUJNOOtOCONâOHCOiOt-OOOONMtOt-Ot-NOOOHOOiO^CO
!-> OtDINCOWHHOOOOO<CM(MCi3NIMHHOr|iÔHna^(MO>(»aiOma>
o>ooa>ooaiaiccoioio>oo«5t-inc-t-t-t-t>t>t-t-ooo5ooaiHOT-it-oooo«)t>
HMio(oaioocotD?oiot-i>Moinooiooot-cO' |
p
j 500C10(C»Offii05DT)iOOait-<0<OOOiOO-*COa:T)i!fl»H«DHNOONMiOOOiO
rh ooooooaoHoot-ootooot-oooooooooioioooiaioNHiMooaiooooco
Oi0t-(0O«0t-C0iO00NOOWOOC0OOONNNOO00O -t-ocoooo
rri OOTi'TfTji03t-t-MOOai0500a)WCOOOOOO(M(N'*'^HH«OiO 00 CO
lOC£>«5lO«5lffllO?£>t-iXilOCO y3COlOlO-^ -^ lO'OlO-^<mcOVOÔOO
|
t- t*
H «5
,
00 00
LO CO >*
< • t£>
ooaocororococooscocooooocooococococococococoeococococo 'cocococococo
00Mii5NNt-«MC0O(0l-OOONMOOOrlii0t-Ot-N00 • C- O t- lO O t-i
O50505050i05ooo5oo505ooo5ooo505ooooo505ooo50505050505 • 05 05 os 05 05 oo
C £v
'"
3 t-'tTfiflOO«n0iO5JHl»M(M'^'*00HHTJ<Ot-'^'<*C0inn00)t>HC0Ni0t-
O O T-l T-l (M T-l
05 05
03
05. 0)
0.33.5 •
Q.33^ ;
o s 0) O 33 a) a)
0)
C/2^ S3
g t-
CQwrH
-^^_ 305
=
03
OP PQ^Q
.S'S s^
« 2^ S
c<J •
£3'
03 03 2 03
S3 >-
S3 S3
Oro PhPh SPhW
215
OiW'<Ji«30J5001WHM(NOOOSt-Ot>000>«)HHOtDt'IN«5NOOiOit>tO
O COOlrHOOlt-COMt-OSOHOXC-MOOt'UJ'^t-t-OOIMOOrHONNNOiOlCOOOOl
coco^^coco-^cocococoeococococococococococo-'tfco^^^^'i^cococococo
OOOH01^Tf{i5COOOMOMO'^HO(DOinOi'<fOOOHiniOtDt-inO'^t'0(M
O H[-NWN'<jiÔ>Nm(MH«)OOHOOONOO>HH^HHtCCCMtDOJ'*OON'^
O COHHNO«INO(MNIMHNOHHOQO'*OrtHHM1010MC003aiMa>(MOT
,/)
C>JCMCMC<lCMCMC>aCMCMOqCMCMCMCMOJCMC>]i-<CMC<lCM<^C^
00 rj< O i-h 00 O U5 O CO 00 lO i— I
LS) • CD CO OO • C- CO OS tfi OS •
t- OO GO t- i-i t- O CO 00
j5
"^ to moo i-h oo t~ oo oo as t- oo to id
eo co co -^ co co co co co co co co co
-toost-co
co co co co
• co oo oo io >*
-cococococo
• oo t- oo co co »o co t- oo
• • co co co co co co co co co
.,-. oooa>^t-(N(Niot-ooô -oocooo •cuoioiONioffloioomocoxo

2 OS
"^
jn-> t> as as as as oo as as oo t- oo to t- -t-oot-co t- oo oo n< io co co oo as as co co to oo oo
co co co co co co co co co co co co co • co co co co • co co co co co co co co co co co co co co co
j 00 r}< CO CO OS CM C- U3 CO O -<tf O -<* OHOSN •

t- OO i-h <M CM CM <M i— to lO asI o to as CM
^
f? t- »o as o
as t-oo oo o
as as to t-
co co co rf co co co co "* co co co co
t-oot-c-
-cocococo
•
•
t- oo as to t- t- oo as as t- lo t- t- oo as
co co co co co co co co co co co co co co co
PL,
-
o Q, as m o o t- oo ^ oo co to t- co t-h -ot-aso -tot-MH • co oo o co as i-i cm i-h •*# co
S3
o i
W oo
CO
o o o as oo as as o as as to t-
CO CO CO CO -^ CO CO CO CO
T)< T}< rj<
-t-oooooo
-COCOCOCO
-ooasot-
• CO CO -^ CO
•
•
oo t- oas as to io oo as oo
CO CO -^ CO CO CO CO CO CO CO
<
c too"<#coasi—tcoasotMioasastotocoooooto-^cMiocMt-LOto-rfLOioioioi— i as to
»-j
O •^fCO-rtuCCMCO-^CO-^^fCOCMCMOCOi—
w IMî-lHIMM^r-l-^-^-^COCOOCM-ÔlO
« cocococococococococococococococococococococococococococococococococo
o
&H
Mt>toast-ooNoot-HiNt--<ttoint>cOt-(oosoc<iioosHiniHcotDc<itOr-(coTf
hJ
a HOHHOoiooHoot-o)t»ooaiffioitr-»oocoaioooHHt-MO)oai
w ^^^^^CO^^^H^êOCOCOCOCOCOCOCOCO^^COCO^-^-'tfTl'^t'COCOCO^t'CO
. cc^toasiocMC<iast^cMtoc<iocMtoooooasO'-icorfcoascocoo>JOtococoiotoo
OS
o Tl'mcOiOMOTTfCOCO'^COMCOOCOHCC'NIN'-iNCOrtlrHTfiOiO'^HON'^rHCr)
w cocococococococococococococococococococococococococococococococococo
^ t-»-ioiotoiooocoooascot-aso]toasuooooooocoT-iast-oo -icotDoocMt-oto I
Hi-icMTHoasasooasoooast-ooasoot-ooasasasooi—iooHHt-ai»oa
o ^^^^^COCO^^CO^COCOCOCOCOCOCOCOCOCOCO^^^^-'tf-'tf-itfCOCOCO'tfCO
w
CMCMcotoast-cMtoasi—iimiOi-ii-iooôoi- uo oo t- co i-it—loooascMoooa-^toco i
CO
2 Wl010^Tf'<fTt iOlOCOlO«)(fl-^Tl COlOOiaOT'^ -*CO'^t010lOtOlO(M»OMTtiL'5
| l <
OJOaC<ICMCMCMCMCMCMC<lC>aCMCaCMC<10aCMCMCMCMCMcMCMCMCMCMW
: o.
s t-Tf*ioootoioioo>HoocoN'*"*oOHHTfOt>rf'*Mio«;a>t-H05c<iiot-
;
'
S
m
i-l H l-H CM
2;
a ^^^^^^^^^^^ÔfOi-Oi-OÔi-Oi-OfOT-OfOi-^^^^^^Ôr-Oi-CM-OfO*-
CD .
HH
o^c-ifli— i oo us oo o ^ t- oi «o 10 r-i-î—i-^T-icôo-ôot-coôo-^t^coc^coasai
£ ONOt-HHOio5Mmwot»^ômio«c^T)i(Moooooo«iab-inoocoo
pj t~t>M'»JiOOO«0«5tOO<OiOCOWOTjiiM'^C00005000t-«)TfTliC]OOC!«DOO'^!0
W 0501r-i«005000COONHO>10«mONM50N(N001WOiHOMOOWTJioOCOO
oit-ic^c<ic»t^co-^<o^t-t-oo-ri<^j<coc<iuD-^<iocoTj<i-Hcoaiixia50cocoiocc>oo
P3
^ HOOO(MO)MOWt-Ot-HOHOlOON10T)<0000(MTii®H«aiNOOOO(010«)
rH.
£— t— CD t*- CD CO C*- CD CD t— CD t~" C-- CD tr- CO t— Cr— CD CD CD CD C— CDC— t*-COCDt~C~" CD C— CD C—
,_, NMt-05C0INHH»mcqTli0>Nt-OWi0OC00lTl'HOO(MM CD OS W Ol Ol t-
o
O H0)O<*OC0i0NMN(NHO0iNrtmi0'J
CDiOCOCOCDCDCDCOCDCDCOCOCOiOCOCDCOCDCDCDCDCDCOCDCOCOCO
l
MOMHOMHO -t-OBHOH
-kCCDiOt— COCO
00O(MOHi(5NN(J>»O050>tCiOHTf00O05Tf(I>iO00Mt-Ot0OO(MTl<t-i0
J HiOWHcgHOrtOOONi'OWMOTfOHHOHNNHOHOOONWO
I I I I I I I I I I I I I I I I II
•^Tj<cOI>W5iO(Ct-OONt-OOOH^05iOHOiOaia!(MOiH«00>C100Ntr-(00
m (MCO-ÔOCDt-airH00050imW05ini001-<*Ot-C~OIMMtOC-OiNOOO«OOOw
w oaoixosao>oo>C!WOi3)a>a!a>ijiC!OOiOiooooo50>ooooc!coo
^lOWOJOlOlOiCOt-lOt-tCHOIMCCiMOOiOOTliOOiOoOCOtflOOOJtDt-NHNW
C-C-C-C-C-C-C-C-C-t-C-C-C-C-t-C-C-OOC-C-C-t-L— C-C-C-C-C-C-C-C-C-C-C-
owt-oi'^NtflinoooO'jc-oincoMHoiOHoooiHt-âiooit-t-cotDt-i-i
J 000[-OTIMIM(niOlMT)iTlic-Cl^(M(Ml(5t-05^'<l lOOOinr-IOH«!OC]'*'ÔOlO ,
t-t-t-t-t-t-c-t-t-t-t-t-cDt-t-t-t-t-t-t-t-c-t-tâjc-t-t^t-t-t-t-cot-
t-ocviot-t-ooco-'tfcDoasoicD'-it-cDOiOT-io^cDaiiooo-rft- co w t- w w a>
coooc-co^j'ioO'—lo-ôococococooooookoo-^coaiooooaico •NHOOffloo
hH
OSlOtDMC-IM(MONTfHOOOiOcgO)inOOiCOOWH(C003t-CO'*OiOWOtD
"a HNNCOHIMINONNIMOHajHOOOOHHHtOHTliTliMMWOHiNHO
I—IHHr-IHHHl-It— T— t— i—IHrtr-It—It-It—It—It—It—It-Hi—It—It— r-IHHr-
I T-H T-l T-l I I I
0>0«DOOU3-<fM05HOt-lOlN«ia5001>iflNa5MO>a5HOOMCDOO«Oia©00005
—c 3a t-T*'<tiOOO^>WiOOSHOOK)NTjiTlioOHH'rJiot-'*T|iCOiOt001tr-HMN"5t-
8 ^^^^^^^^^^ÔrOrC»C*C*OfC>C>^C*C*^
0)
> >
>t-I(M
D.3-C o>
T3 "S *3) >
<D o C 0> O 3 0) <D
Si
SPH QQ^Q -
it* c3oa5a5£aJc!
"Cu, 3 n
2 SoflrP'C 5S3 S 9-TH C5 2 cS
rt
J 3
-
,3 fc.3.J-j ©."S3 8 P.cU
S»
3 3*2 3
M&&%$&Qmtt& <5 i^ PQ Ow fin Cm S Pn PQ PmPu,t>P4
217
n t-HHt-t-HH(DH0i(MO'*i/5'<li0)HHinffi'*HOO!N»i0NOi0!£lOt-O
ON
^ (NO)COi-lrHO)rHCOU500eOTHCO(MHMO>COOiHOINt-U5C>rH10t-IJ)1005«l[-t-
fi"&,
ooc~oooooot-oooooot-oooot-ccoooot-ooooooc-ooc-t-t-oot-t-t-oot~oot-t-
l0C-r-KNt-»00lN0)a>Clt-t><arH00NrMMOlM00C3OrH0Jt-X!0WCCrHCCO
ooooooooooooooooooooooooc-ooooooooooooooooooooc-ooooc-c-ooooaoooc-t-
NbMNOOWmt-NMOOOiOOOIMt-OOO'^mNNt-NOOOOOt-lOiOci
2n oo^c^co^tfsco^cJôjeo^t^^T^cNcoLra^cocofNiasc^cotfsooîoc-^co^
CDC-t-t-C-C-t-t-C-t-t-E-t-t-t-t-C-C-t-E-t-t-t-C-C-t-C-t-t-t-E-<Ot-t-
Ot-MO^NiOOWOOOOiOOiOOOOiOHt-iOOWtCWOfflOOOmM
03 ^ 00005rHOO<C01C-000>C-GSa>1000t-OOOOlflHOtO[-Ot-rH0500NO)001><Nt-iH
Tfcoco-rt<cccocococococococooococococococococo^co-^<coco-<*cococo-«3<coTi<
"W
. ôo '—
* cm . .o . .
. .05 . "* . . »0 . lO
W 00 OJ lO<Ntfl© oJOoo
>« *-^© JJ o>
-S^ 2
» CO
O >J H .O .1—1 . .O .1
't~co t-
O ^M^W^tO JJ CO CO CO
a
K CO .OS .CM . .CO .O .CO .
~~ ''*LO'-H 00Ô>©©CMcO'— t-MrJ"
'
H
CO
H
V. FREQUENCY DISTRIBUTIONS AND SEX DIFFERENCES
Introduction
Measurements for the sexes were kept separate throughout the
statistical work in order to estimate the nature and extent of the
differences. These differences are shown in a series of curves of fre-
quency distribution (Figs. 12-24).
The procedure has been to plot on each diagram two curves,
one for 124 northern and southern males combined, the other for
70 northern and southern females combined.
The sex differences are shown in tables (pp. 219-221), where
averages are given for 44 cranial characters and 15 indices. Another
table (p. 224) gives sex ratios for 59 measurements and indices.
The sex ratio is found by dividing the average measurement for
124 males by the average measurement for 70 females. For com-
parative purposes I have worked out sex ratios for measurements
given by Pearson and Davin (1924) for a series of Egyptian skulls,
and for other racial samples mentioned in the table (p. 224).
Interpretation of Diagrams
The points to which attention should be given in studying Figs.
12-24 are: (1) the smoothness of the curves; (2) the incidence of
the modes; (3) the distance covered by the bases of the curves;
and (4) the relative positions of these bases on the abscissae. The
method of study may be illustrated by reference to Fig. 12, which
gives data for cranial capacities of males and females:
(1) Both curves are without important irregularities; they both

conform to the curve of normal distribution. The smooth-
ness of the curves indicates that the data are adequate for
plotting and that they are homogeneous.
(2) The modes occur definitely at the interval 1130-1190 cc.

for females, and at 1250-1310 cc. for males.
(3) The bases of the curves indicate that the range of cranial
capacities is greater for males than for females.
(4) The cranial capacities for females touch a class interval

lower than that reached by the distribution for males, but
they do not extend nearly so far into the high class intervals.
The ranges are 1010-1310 cc. for females and 1070-1590
for males.
222
Frequency Distributions and Sex Differences 223
Measurements of Skulls in a Village of Head-Hunters

on the Coast West of Merauke
(Measurements made in situ by Dr. A. B. Lewis)
L B C.I. Remarks
Sex and age (175 125 71.4 M 2 erupted, not M 3
Juvenile J
164 115 70.1 M 2 erupting
166 115 69.3 M 2 erupted, not M 3
{ 168 133 79.1 M 3 erupting

Averages 168. 122 72.5
185 128 69.2

170 124 72.9
195 123 63.1
170 120 70.6 9 male Merauke skulls in Field
183 133 72.7 Museum collection give average
Adult males. 189 130 68.8 L 181.7; B 134.2; C.I. 73.9
165 130 78.8
185 126 68.1 Averages for all male skulls (124)
191 131 68.6 in FieldMuseum collection are:
175 131 74.8 L 177.6; B 131.3; C.I. 74
.
188 132 70.2
Averages 181.4 128 70.7
( 175 125 71.4 5 Merauke female skulls in Field

Adult females J
170 130 76.5 Museum collection give average L
165 130 78.8 178.0; B 122.2; C.I. 68.7
( 171 125 73.1 Averages for all female skulls
Averages 170.2 127.5 74.9 in Field Museum collection are:
L 170.3; B 126.5; C.I. 74.3
175 131 74.8
183 125 68.3
177 121 68.4 Owing to the crude method of
182 128 70.3 measurement, the small size of
182 120 65.9 samples, and the doubt about sex,
181 129 71.3 reliable comparisons with Field
166 130 78.3 Museum series are difficult to
Adults, sex?. 181 119 65.7 make. As one might expect in
188 126 67.0 small samples there is high varia-
175 127 72.6 bility. This variability may be
175 120 68..6 due to the acquisition of these
175 130 74..3 trophies during head-hunting expe-
177 122 68..9 ditions in different regions. The
181 132 72.9 general cephalic trend, from low
176 127 72.1 to medium dolichocephaly, is clear.
173 120 69.4
Averages 177.9 125.4 70.5
There is no necessity to discuss Figs. 12-24 in detail since all

can easily be interpreted by the same method. Fig. 22, showing
curves for nasal indices, is of exceptional interest in indicating the
high variability of the nasal index. The curve for females is ex-
tremely irregular. The curve of facial index for males (Fig. 23)
suggests two types of face differing in shape. The curve has two
distinct peaks, one at index value 66-69, the other at 72-75.
Sex Ratios
Capacity
in cc. L B B' H' LB S, V s2
1 1.110 1.043 1.038 1.053 1.025 1.031 1.038 1.041 1.043
2 1.043 1.047 1.025 1.027 1.038 1.053 1.029 1.037
3 1.063 1.066 1.049 1.037 1.061 1.065
4 1.105 1.038 1.046 1.041 1.054 1.053 1.043 1.045 1.035
S,' BQ' GB J G'H GL s3 8,' s
1 1.048 1.037 1.054 1.067 1.054 1.031 1.035 1.029 1.036
2 1.037 1.040 1.065 1.055 1.047 1.033 1.025 1.034
3 1.062 1.079 1.084 1.063 1.055
4 1.036 1.051 1.072 1.074 1.049 1.041 1.026 1.040
GLU u NHR NHL NH' NB OiR 2 R 0,L
1.047 1.040 1.057 1.057 1.058 1.033 1.036 1.024 1.036
1.038 1.055 1.034 1.031 1.007 1.031
1.052 1.070 1.048 1.031
4 1.038 1.071 1.073 1.065 1.046 1^012 1.009 1.015
Lac Lac
2L o,'R Oi'L 0,R 0,L sc G, G 2
1 1.027 1.031 1.032 1.029 1.032 1.056 1.033 1.061

2 1.004 1.049 1.049
3 l!058 1.067
4 1.009 1.028 1.021 1.032 1.079 1.058
100B 100H' 100B 100NB

G,' EH fml fmb
L L H' NHR
1 1.017 1.154 1.050 1.033 0.996 0.985 1.008 0.990
2 1.040 1.034 1.041 .980 .992 .988
3 1.031 1.045 .984 .996 .998
4 1.071 1.175 1.014 1.020 1.009 1.015 .993 .978
100NB 100 0-,R 100 2 R 100 2 R 100 (B-H') lOOG'H 100 G 2

Ocl
NH' C^R 0,'R LacOiR GB G,
0.983 0.993 1.016 1.016 0.990 0.997 1.015
.991 .972 .993 1.015 .999
'
.975 !974 .983 1.025

.983 ^997 .981 .985 0.934 .976 1.021 !983
100G, lOOfmb Pros

G AA Nz. PZ.
2 fml PL
1.026 1.090 0.984 0.996 1.018 0.993 1.012 1.000
1.006 1.004 1.005 0.992 1.010
1.013 0.998 0.991 1.012
0.989 1.103 1.006 1.004 i!625 0.980 0.998 6^998
1 = Field Museum Series (New Guinea).
2 = Egyptian Series, Pearson and Davin (1924).
3 = New Britain, von Bonin (1936).
4 = Hythe Crania, Stoessiger and Morant (1932).
Conclusions. —
Examination of frequency distributions, for traits
recorded separately for males and females, indicates that the sex
characters are clearly differentiated with regard to ten out of twelve
cranial characters considered. Males have decidedly greater cranial
capacity, have a longer and broader skull, and show greater doli-
chocephaly than females.
Frequency Distributions and Sex Differences 225
Males have the greater cranial height, also a longer vertical arc.
The arcs over glabella, over ophryon, and from nasion to opisthion
(sagittal arc) are appreciably greater for males.
But in three respects, namely, nasal index, facial index, and
facial triangle, the difference between the male and female popula-
tions is not so well emphasized as in the other sex comparisons.
In general, the figures clearly impress the desirability of treating
statistics for males and females separately. The sex differences
are so noticeable as to suggest that compounding of figures would
give results that are not truly descriptive of either males or females.
A pooling of figures would be strongly influenced by the relative
proportions of males and females, and preponderance of one or the
other sex would lead to appreciable divergence in estimates of the
racial characters as shown by average cranial measurements.
Measurements Made in situ by Dr. A. B. Lewis

The shown on page 223 gives a few measurements made on
table
west of Merauke. Dr. A. B. Lewis,
skulls in situ at a village a little
who recorded these measurements, was obliged to use carpenters'
callipers, and the distance between the limbs of these was measured
on a linear scale graduated in millimeters. Plates XXXVI and
XXXVII show the skulls measured in this way. The table includes
Merauke data based on the Field Museum collection. These data
are compared in the "Remarks" column with figures recorded by
Dr. Lewis.
_1_
"930 ,„^„*iOO /P20/2 8(j/340 /4DO^

/4-60
' /3BO >
/04-0 '/S20- ,
Fig. 12. Cranial capacities.

226
Fig. 13. Lengths of skulls.
227
Fig. 14. Breadths of skulls.
228
22
20
IS
,6
\
<4-\
12 \
io\
o}
G* ee ee y
Fw- 15. Cranial indices.
229
119 123 J27 131 135 139 /-^3 1-4-7
Fig. 16. Skull heights, basion-bregma.
230
Fig. 17. Horizontal arcs over glabella.
231
Fig. 18. Horizontal arcs over ophryon.
232
Fig. 19. Sagittal arcs, nasion to opisthion.
233
Fig. 20. Vertical arcs over bregma.
234
106 109 /l 2 //S //8 /£/ /2^t- /Z7" 13 O /33 /36 /39 1*4-5
Fig. 21. Bizygomatic widths.
235
%, ^3 45 4-7 4-9 S/ S3 &S 57 &9 6/ 63 6S
Fig. 22. Nasal indices.
236
G*- sr eo e3 ee 69 72. v^ tb ei e«f-
Fig. 23. Facial indices.
237
^ /
GU
Northern Groups
GL.
Southern Groups
Fig. 24. Facial triangles. Male in broken line, female in solid line. Scale 5:6.
238
[>ur
ills
THt UBHAHl
OF THE
Average Horizontal Contour,
124 New Guinea Male Skulls
Average Vertical Contour,

124New Guinea Male Skulls
VI. CONTOURS OF SKULLS FROM NEW GUINEA
AND OTHER REGIONS
In order to give a quick visual impression of the contours (back,
top, and side views) of 124 New Guinea male skulls, drawings have
been prepared. Tables on which the contours were based are given
on pages 240-242.
The composite contours (facing p. 240) show back, top, and side
views of 124 male skulls from New Guinea. Each figure has been
made by superposing 124 drawings. Composite drawings of this
kind have the advantage of showing extreme sizes, great and small,
also the densest clustering of contours. Experiment has shown
that a line drawn through the middle of the densest part of the
superposed contours gives close correspondence with the average
type contours (facing p. 238), whose calculation is a tedious process.
Contour drawings (facing p. 242) show the degree of coincidence
between the vertical, horizontal, and sagittal sections of 124 male
skulls from New Guinea; 62 male skulls from New Britain (von
Bonin, 1936); 46 male skulls from the Wateita tribe of east Africa
(Kitson, 1931); and 82 male skulls from Australia (Wagner, 1937).
The drawings indicate a close fit of the vertical contours of skulls
from these regions, and the lines are in places coincident. The New
Britain skulls have a somewhat higher vault than the others and
they are the broadest across the porion line. Between the vertical
contours of New Guinea and Wateita skulls there is practically
no difference.
The and New
horizontal contours indicate that the east African
Britain skulls are similar in length and general shape, and that both,
though similar in shape and size to the New Guinea skulls, are
appreciably longer.
The sagittal contours indicate that east African and New Britain
skulls have a fuller frontal region than those of New Guinea.
The occipital region of the New Britain skulls is almost entirely
coincident with that of the New Guinea skulls. But the occipital
region of the east African skulls is more protuberant than any of
the other occipital contours. The alveolar, palatal, and basion-
sphenoidion lines are close; so also are suborbital points, two of
which coincide. The positions of nasion, glabella, bregma, and
lambda are close, so indicating similarity in the size and shape of the
longitudinal arcs.
239
OS CJ >< COCO I <M
(D CON o*
co co co
co CO
in t-
03
& ^*
co
COCO
•
CM t-
to ^ CO
CO CO
ours
> Norther
nea SkulW
r c
a
'7
33
mw
<
H
3
<
CO
a
5
P
o
£
es
o
z
o
o
<?
H
Z
o
h
p
o
H
Z
O
u
<
H
Z
O
Ne
W
in
in<
THE UCSARY
Of THE
UNIVERSITY OF ILLINOIS
Test of Random Choice of Traits for Working Out a Coefficient of Racial Likeness from Skull Measurements of Males from Northern and Southern New Guinea
Crude
Designation Numbers of Traits Selected Are Given in Italics at the Heads of Columns in Tables, pp. 266-269 v. CRI
Av.'cRL
3.52
5.03
4.34
6.37
3.92
4.87
4.92
3.52
5.45
PE of each CRL when 30 traits are considered is ±0.04. Av. = 4.63±0.1

Crude
Av. CRL
25 27 5 42 46 11 41 37 30 24 46 18 29 12 48 10 47 49 33 3 2 14 15 4 52 34 2.97
25 56 23 50 13 28 19 4 53 41 51 55 48 10 33 3 59 20 27 52 11 34 2 17 25 29 5.31
PE of each CRL when 25 traits are considered is ±0.05.
•in 60 48 11 64 53 29 23 45 46 41 42 15 3 39 IS 24 16 34 35 47 4.04
20 57 28 37 65 40 33 44 58 48 38 29 27 22 11 7 52 12 14 17 4 6.74
20 27 60 32 49 20 30 51 34 1 56 6 44 58 42 47 53 65 11 40 52 6.31
20 16 40 37 47 16 41 31 36 44 10 53 14 7 29 6 54 27 2 56 3 3.87
on 48 24 23 2 27 9 7 36 42 55 54 3 16 12 47 50 39 45 11 29 4.63
20 37 65 44 33 36 38 18 54 16 1 46 4 12 16 10 32 39 6 43 7 5.61
90 49 1 64 12 16 10 34 41 15 5 4 40 22 55 44 56 32 62 3 2 5.65
24 60 31 10 39 9 42 13 35 57 19 36 17 34 47 3 44 55 28 33 6.95
20
PE of each CRL when 20 traits are considered is ±0.05. Av.=5.26±0.16
50 26 6 27 24 9 13 47 12 30 7 32 19 46 2 2.58
51 23 36 11 46 6 59 40 53 9 49 44 26 31 48 4.37
f each CRL when 15 traits are considered is ±0.24. Av.=3.92±0.
Crude
Av. CRL
56 48 46 6 43 41 36 33 37 1 4.21
50 9 27 28 16 21 20 47 53 13 2.92
19 63 51 23 65 7 31 47 1 68 10.85
60 42 12 21 38 48 19 69 10 57 5.40
8 22 9 5 46 24 41 62 33 4 1.92
15 46 47 21 39 3 13 25 22 40 3.99
1 36 46 16 26 14 44 54 48 43 4.51
29 3 41 26 50 14 48 62 65 39 5.07
2 55 32 69 44 11 28 13 48 52 9.36
58 34 31 12 9 1 54 53 30 43 6.92
f each CRL when 10 traits are considered is ±0.28. Av. = 5.51±1.50
Contour Studies 243
The technique of drawing the contours and of superposing

them was taken from Benington and Pearson (1911), Tildesley
(1921b), and Morant (1923). The style of my contour tables is
based chiefly on those of Kitson (1931, pp. 305-307), with minor
modifications. When superposing, the essential requirement is to
have exact coincidence of the vertical axes, the horizontal axes,
and the points at which the axes unite. The essentials of the tech-
nique have been briefly summarized by Wagner (1937, pp. 35, 36).
The instrument used was the "Universalstereograph" made by
R. Schwarz of Basle.
In a later chapter (pp. 264, 265, 266-269) a statistical study has
been made of Negro craniometry in relation to that of Australia, New
Guinea, and New Britain. These contour drawings give a good
visual impression of the numerous resemblances and differences
expressed by the table (pp. 266-269) of averages with their A/P
A
values.
VII. COMPARATIVE STUDY OF CRANIOMETRY OF
NEW GUINEA AND ADJACENT REGIONS
Introduction
The monograph would be a thorough
desirable conclusion of this
comparative study of Field Museum New Guinea samples with
skull measurements taken by different observers in other parts
of the island. The inquiry might then be extended in order to
compare cranial measurements from Australia, Melanesia, and
Africa. Inclusion of Africa in the study is justified by the fact that
there is apparently a strong Negro element in the Melanesian area.
Such a study has indeed been attempted by Wagner (1937),
who gives a large bibliography. His work is important enough
to merit consideration in some detail. Our review must relate to
two aspects of the problem: firstly, a consideration of general re-
semblances between racial series by use of the coefficient of racial
likeness; and, secondly, a study of average measurements for racial
groups of skulls. We shall have to take cognizance of many statisti-
cal difficulties.
The Coefficient of Racial Likeness

The points for discussion under this heading relate firstly to
the nature of the coefficient itself, and secondly to the specific appli-
cation of the coefficient as made by Wagner in his comparative
study of the Oceanic races.
Probably all statisticians are agreed that the concept of a co-
efficient of racial likeness is desirable for expressing in one figure
the sum total of the differences and resemblances between series of
measurements made on two different groups of skulls. But the
following quotations indicate that not all statisticians are satisfied
with the coefficient as a valid means of comparison.
Pearson (1926, p. 105) states: "The CRL should be a measure,
not of how far the two races or tribes are alike or divergent, but
of how far on the given data we can assert significant resemblance or
divergence."
Fisher (1936, p. 60) states that the term CRL "suggests that it
affords a measure of the differences, or inversely of the likenesses
between different races. it does not attempt to do,
This, of course,
nor is any needed for doing this; for the
special statistical device
differences between the averages of each particular measurement in
244
Comparative Study 245
the available samples afford the most direct estimates which we can
have of such differences as may between the populations
exist
sampled. CRL is defective in that it takes no account of the cor-
relation, or covariation of different measurements of the same skull,
but treats them as if they were statistically independent."
The criticisms of Seltzer (1937) relate to what he calls construc-
tional defects in the formula, theoretical defects, and inconsistencies
in the results obtained.
To these criticisms von Bonin and Morant (1938, p. 96) have

replied as follows: "This method has recently been criticized with
little regard to the fact that its limitations and imperfections were
fully recognized by its inventor, or to the way in which it has been
used in practice for more than 10 years. For practical purposes,
the crude coefficient of racial likeness remains still the best means
to estimate whether two samples may be considered to represent
the same population or not, and the reduced coefficient remains an
effective criterion of the presence or absence of a racial bond between
two differentiated samples. Past experience gives no reason to
believe that the method of the coefficient of racial likeness fails to
provide close approximations to the results which could be obtained
by applying theoretically more correct formulae, such as those taking
into account all the intercorrelations of the measurements used,
but which have the disadvantage of involving many times as much
arithmetical labor. The desirability of using a large number of
characters has repeatedly been pointed out. But the method of the
coefficient of racial likeness —
being admittedly a 'stop-gap' is —
not a simple rule of thumb. The way in which its values have to be
interpreted in order to yield useful results has to be determined
empirically."
The main criticism of the use of coefficients of racial likeness
for comparative study has centered about the number and nature
of the traits which should be chosen for representation of the racial
types.
Wagner (1937, p. 122) points out that coefficients of racial like-
ness which are calculated from a differing number of characters
should not be compared. Measuring techniques should be identical
or almost so. He notes truly that even if 20 to 30 traits are measured,
some few characters can greatly affect the CRL. We shall see later
that Wagner considers the CRL of some exceedingly small samples,
and that the number of traits is often whittled down by him to as
few as 14.
But without further emphasis of difficulties let us consider the

matter experimentally by detailed analysis of my New Guinea data.
We are evidently at a stage in which each worker must squeeze
from his records the maximum amount of information. My test
by itself will not settle the points of dispute; yet the pooled results
of many workers would help to solve our statistical difficulties in
an empirical way.
Referring back (pp. 127, 128), it will be found that the crude CRL
for males of the northern and southern groups was 3.60 ±0.17 when
29 traits and indices approved by the Biometric School were selected
as a basis of comparison. The crude CRL for females was 2,90 ±0.17.
The following table includes the result of basing the comparison on
the 30 traits and indices not preferred by the Biometric School, and
the data also give the coefficients of racial likeness that have been
calculated from all 59 traits and indices which are available.
CRLValues for Males and Females of

Northern and Southern New Guinea
Traits chosen Males
1. Based on 29 standard traits
(Biometric School) 3 60
. ±0 . 17
2. Based on the other 30 traits . . 5 97 ±0 04
. .
3. Based on all 59 traits 4 81

. ±0 . 12
Reduced coefficients are in bold-faced type. The probable errors

have been calculated from the expression
0.674489^(1-^)
where M = the number of traits used for comparison.
A/P A Values for Crude Coefficients of Racial Likeness
Males and Females of Northern and Southern New Guinea
Comparison of Males Females
1 and 2 13 62
. 8 62
.
land 3 5.82 3.65

2 and 3 9.21 5.87
The experiment shows that significant differences in the values

of the coefficient are obtained by changing the number and character
of the traits selected. But
is evident that great latitude can
it
be allowed in the choice of traits since the ranges of the coefficients

are only from 3.60 ±0.17 to 5.97 ±0.04 for males, and still less
(1.40 ±0.04 to 2.90 ±0.17) for females. By varying the number
and identity of the traits we have indeed produced a mathematical
change in the CRL. But the change is not of great practical im-
WNeONO«3^H«;MH!00«0>0
Ht-MHOMOsooMooooi-iat-om .. HHoooomMHTfNoocooHoxxiin
P3
J HNOON«DCOt-5Dt>t-HOO>o:00
M t-N(Ct-IN(M00WOaiHOt-0)H00
OHCOOOOr)<000'*50r-lM'*OOr-(OOOS
i— rrs co oi
i-yii— nn 10 to m fo
OiKHCOIMNOOOOOUJt-WNmOlM
<-~s fvi > rv^ r~- rvi *vi
iOOt-OONTfOHt-MN(Ot-00 c
tf
,_,
«oocooooNMa>HO»ioa)HHH
XWWMin(00000(MO<C<OHOiO
moisooiMc^cooôooocot-coio MiO«OHMt-INNOU50iOiOt-(CO
£ r-H i-l CI
CM 1— 1-1 OS i-H 1 CO <M i-( NN^
u
U 1
£ .5 3
g a o o as O
!T3
, C ccj "C c S c
< g.-'PQ - o .J CO o; 'C s O
cci
W jS.Jffl O cW ^'-
co co is > §|T3 O £
3.
,
2"K 2 &
^ £ fin H<5 §ZO.
H3 TJ T3 T3 531 T3T3
c c c c c2
cd w ri Tt c« C
c c c "O T3 T3
C 3
Ti 7i 7t 7Z
05 Dj
C C
^ -* ^c *>
rt a! ctf ccj c3 £3
a a cô
C3 Jfi
ai a> <D a> ^

8 rt c«
ca
O
rt
Qi
ctf
<v
rt
a>
crj
a>
eS, rt rt rt
^3 ^3 cS
3 3 flflS. T3X)
.£ 5 £ 2 v, u AC'S cj rt c3 C G
•|oooo
3«S t- C3 C3
f-i.2.2
^S rt
C3S fci
J33
•|oo
3 3 3 t, >-, tn J-
^S2
.2.2.S c3 as
cq rt
H *- tn (h'C'C
o J* & & & +3 +3
CO
t-i
CO CO
hi »-.
§ £ Es ^ & & ^co ^co -u H;
g g g § §
CO CO
BlcJcJ 33 sill's a; a> <x>
co
www
247
w co -«s" co as cd ww as t- cm co co to o a-.
OSOCOCOCOCvl'tfCOCO-rr^Ht-asCOCOas
eg co cm i-i oo t-i co ** co as as Tf as as
inr-HDlOlOC50Jr-IOlOr-lTH(X)l>COlO
PQ
O J
.
I
CD H CO CO •** t-h i— i— CM CO
I I
**• 00 GO i—
I OH ^•OCO-^t-t-COCDiOOCOOOOOlOt-Oi
CM i-H
<!NJ
OCMCDCOOOOOCMCDOOC'ÔOIOCOt-i
NWOOTji^r-li-iO-^t-COL-lOCCT-HC-]
portance, since the CRL values continue to show very close resem-
blance between the two series of skulls.
Now it cannot be denied that the kind of traits selected for
comparing two racial series must make a difference to the CRL
value. This point was dealt with by Morant (1928, p. 336), who
gives a table of alpha values for each of 31 traits. The alpha values
are based on a large number of observations; the number is given
in Morant's table. A low average alpha value for a pair of traits im-
plies uniformity of measurement in the samples used. Conversely, a
high average alpha value indicates great variability for those traits.
Morant's experiment indicated that some characters, because of
good as distinguishing features when
their uniformity, are not so
racial differences are the object of research. The cranial index,
the breadth of the skull, and the height index are three of the best
out differences in two craniological series.
traits to use for bringing
The index foramen magnum tends to uniformity, and is,
of the
therefore, not so good a trait to use for comparison of two series
of skulls.
The (p. 250) was worked out from my

table of alpha values
comparison of from six regions (pp. 247, 248). Since
series of skulls
there are six regions, we have 15 comparisons, for every possible
pair has been considered. The number of traits used for comparison
was 32.
On the whole, G 2 (palatal breadth) is an unreliable character to
choose for distinguishing between two series, and B' and 100 fmb/fml
are not much better. The N
angle ranks high as a distinguishing
feature, as also do the breadth of the skull, the horizontal circum-
ference above glabella, the nasal index, the cranial index, and the
height-length index of the skull. numbered from 16-32
All the traits
would serve well as distinguishing features when comparing two
series of skulls. In 9 out of 15 racial comparisons B' fails to give
any distinction; but the nasal and cephalic indices are distinguishing
traits 13 times (see AP A values, table, pp. 266-269).
There can be no doubt that, if given a choice, some traits will
serve our comparative purpose better than others. But suppose
we do not have much choice, and have to accept the number and
kind of traits that happen to be on record; in what degree will the
crude CRL be affected?
An attempt has been made to answer this question by a further
experiment with my New Guinea data. The results of the test are
given in a table (facing p. 250).
List of Averages of Alpha Values

(Based on fifteen groupings, taken a pair at a time)
(1) G, 8.38 (18) J 40.51

(2)B' 8.56 (19) BQ' 42.08
100 fmb (20) Capacity in cc 42 72
.
(3) 12.07
fml (21) AZ 43.79
(4) 2 R 15.36 (22) L 45.61
(5) G'H 15.58 (23) GL 51.29
< 6 )° cI 15 73
-
(24) G, 51.44
(7) NHL 18.80 „,
(8) LB 20.30 (25) 52.70
(9) S 21.45 L
(10) BZ 24.94 NH'
—
(26) 53.85
(11) H' 29.01 100B
(12) fml 30.21 (27) 56.15
(13) fmb 30.89 L
100 G 2 (28) OiR 59.18
(14) 33.55
G, 100 NB
(15) NB 33.75
(29)
~^T 79 73
'
(16 ) Gl ' 39 03
-
(30) U 80.38
10 ° 2R
(17)
°
39.48 (3D NZ 99.19
OiR (32) B 119.63
The namesof the traits were written, each on a separate piece

of paper, and then drawn from a box at random. First of all, the
trait slips were drawn 30 at a time, and after each drawing they
were replaced in the box and well shaken.
The table shows that 10 drawings were made of 30 traits at a
time, 10 of 25 traits at a time, and so on down to 10 drawings of 10
traits at a time; thus 50 drawings were made. Since each trait
has a designation number (table, pp. 266-269), a list of the traits
drawn at each trial was easy to record by means of these numbers.
is the CRL value for each of the 50
In the right-hand column
draws. Our experiment
not concerned with the significance of
is
small differences between each pair of CRL values. We are mainly

concerned with the general changes in CRL value when we vary
the number and nature of the traits. The changes in CRL values
can be studied by comparing the ranges and the averages of CRL
values for the five groups in the extreme right-hand column.
So long as we keep to 30 traits we can vary them widely, and

the CRL will range only from 3.52 to 6.37, with an average value
of4.63±0.15.
No.
-4
Cutting down to 25 traits makes hardly any change in the CRL,

and at the same time we have tried 10 new combinations of traits.
The new range of CRL values is from 2.97 to 6.62, and the new
average is 4.68 ±0.30.
Since we have five averages there are ten ways of comparing
them two at a time and meanwhile taking their probable errors
into consideration.
A/P A values from Mi-M 2 /V(PE,) 2 +(PE 2 ) 2 are
0.15 0.99
2.86 0.54
0.99 1.87
0.58 0.17
1.70 0.96
This experiment suggests that, provided the samples of skulls

we may vary the number and nature of the traits
are fairly large,
considerably, and still obtain a coefficient that indicates a close
resemblance between the two series of measurements.
If, however, we turn to the literature quoted by Wagner, and
consider his specific use of the miscellaneous skull measurements
presented therein, our real difficulties of comparative study are
apparent.
(1) Many small samples are given, but often we do not have
adequate samples to represent an area.
(2) In order to obtain an adequate sample from any given area
we enter upon the dubious process of "pooling" measure-
ments made by different observers. Even with the utmost
care the process is sometimes left open to criticism.
(3) Some observers measure very few traits. We really ought

to have about 30 carefully selected traits in each of the
series under comparison. The data in general yield nothing
like this standard.
I present with some diffidence an analysis of 244 crude coefficients

of racial likeness (see table, pp. 252, 253). Some of these were kindly
supplied from the unpublished data of Dr. G. von Bonin, some are
of my own calculation, many are from Wagner (1937).
A few general trends, on which I would not place too great

emphasis, are apparent; this is so, despite certain valid objections
to the inclusion of some coefficients which have been worked out
from samples that show one or more of the defects we have just
discussed.
Analysis of CRL Values
CRL Values
3 10
Regions Compared to to to
3 10 20
Africa (Wateita)
—
Africa New Guinea 1
—New Britain
. .
Africa 1
—Australia
. .
Africa 1
Australia (Internal) 35 6 2
Australia and Melanesia
—
Australia New Guinea 8 3
Australia — New Britain 3
— Solomon Islands (Santa Cruz)
. .
Australia 1 1
Australia —New Caledonia 4 3
Australia and Polynesia
Australia (A) —
Tahiti . 1
Australia (A)— Loyalty Islands 1
—
. .
Australia Loyalty Islands . . 1

(Wagner-Hrdlicka)
Australia — Maori . 1
Melanesia (Internal)
New Guinea (Internal) 7 23 4
New —
Guinea New Britain 2 5
—
. .
New Guinea New Caledonia 1 1

New —
Guinea New Ireland 2 1
New —
Guinea Solomon Islands (Santa Cruz) 3 1
New Britain — New Caledonia . 1
Britain — Solomon Islands (Santa Cruz) ....
.
New . 1
New Britain — New Ireland . . 1
New Caledonia (Internal) 20 7
New Caledonia— Solomon Islands (Santa Cruz) . . . 1
New Ireland — Solomon Islands (Santa Cruz) .... 1
Melanesia and Polynesia

New Guinea — Loyalty Islands 2 1
New Guinea —Tahiti 1 1
New Guinea — Easter Island . . 2
New Guinea — Maori 1
Caledonia — Easter Island
. .
New 5 4
New Caledonia — Loyalty Islands 1 1
New Caledonia — Maori . . 1
New Caledonia — Tahiti 1
Britain — Tahiti
.
New . 1
Polynesia (Internal)
—
Maori Cook Islands 1
—
Maori Society Islands 1
—
Maori Marquesas Islands 3 1
—
Maori Sandwich Islands 2 . . 2
—
Maori Moriori 2
—
. .
Maori Easter Island . . 2

—
Maori Loyalty Islands . 1
—
Marquesas Islands Hervey Islands 2
—
Marquesas Islands Tahiti 2
—
Marquesas Islands Society Islands 1
—
Marquesas Islands Moriori 1 1
—
Marquesas Islands Sandwich Islands 2 2
—
Marquesas Islands Cook Islands 1
—
Marquesas Islands Easter Island
Analysis of CRL Values — continued

CRL Values
10
Regions Compared to to
3 20
Polynesia (Internal) continued
—
Society Islands Cook Islands 1
—
Society Islands Moriori
—
Society Islands Easter Island
—
Sandwich Islands Society Islands 1
—
Sandwich Islands Moriori
—
Easter Island Cook Islands 1
—
Easter Island Loyalty Islands 1
—
Cook Islands Moriori 1
—
Cook Islands Sandwich Islands 1
Tasmania
Tasmania —Australia (West) 1
Tasmania — New Wales
S.
Tasmania —Victoria
Tasmania — Australia
(Morant's A series)
Tasmania — Queensland
Tasmania — Australia
S.
Tasmania — N. Australia
Tasmania—Victoria
Tasmania — New Guinea (Watam)
Tasmania — New Guinea (Kai)
Tasmania — New Guinea (Onin)
Tasmania — New Guinea (South)
Tasmania — New Guinea (Papua Gulf)
Tasmania — New Britain
Torres Strait
Torres Strait — Australia
Torres Strait — New Guinea (Watam)
Torres Strait — New Guinea (Humboldt Bay)
Torres Strait —New Guinea (Geelvink Bay)
. .
Torres Strait — New Guinea (Papua Gulf)

Torres Strait —New Guinea (South)
Torres Strait —New Guinea (Kai)
Torres Strait — New Guinea (South West)
Torres Strait — New Caledonia
Torres Strait — Tasmanians
Racial Elements in the Pacific Area (for detailed tables of co-

see Appendix).
efficients of racial likeness, —
The general facts of
Australoid and Polynesian migration are summarized by Wagner
(1937, pp. 7-24), and the data reviewed in my summary (pp. 259-
261, 264, 265) indicate how typical Melanesian measurements have
been here and there modified, especially near the border of Polynesia
and Melanesia. Our research will give statistical support to the
generally accepted views concerning resemblances between African
Negroes and Melanesian peoples.
Perhaps, if we studied more closely the prevailing winds and
currents of the Pacific Ocean, the apparent "hit or miss" of certain
intrusive traits might be better understood.
That the migrations and mixtures have apparently been some-

what haphazard is illustrated by Figs. 25 and 26, showing inhabitants
of the Gazelle Peninsula of New Britain. The photographs were
taken by Dr. A. B. Lewis, who points out that the Baining or hill
people (Fig. 25), who do not speak a Melanesian language, are
definitely of African Negro appearance. On the contrary, their near
neighbors (Fig. 26) are coast dwellers of Australoid appearance.
In this studypropose to work outward in each region from a
I
consideration of intra CRL's to those which may be called inter

because they show mathematical relationship between series of
figures from widely separated areas. Only crude CRL's are here
considered.
have shown that a CRL below 3 expresses a very close sta-
We
relationship between two series of averages.
tistical Moreover,
a CRL from 3 up to 10 (according to the number and nature of
the traits selected) is a reliable indication of close relationship be-
tween the two series of averages, compared trait by trait. We can
also safely say that if the CRL is not more than 20 the relationship
of the two series of averages is sufficiently close to arouse our
interest in the possibility of ethnic relationships.
We will not consider CRL values above 20-21, though some
high coefficients (40+) are derived from two series of averages in
which as many as 8 out of 32 of the averages show no significant
difference. We may speak mathematically of first-, second-, and
third-degree relationships; these are suggested by the CRL values
0-3, 3-10, and 10-20, respectively.
The geography problem should give no trouble since I
of the
have indicated (pp. 274-281) the connection between minor localities
and the well-known names of large groups of islands, which are easily
identified on any map of the Pacific. The Times Atlas is very service-
able for reference. Society Islands is an old name for the group
which includes Tahiti. Sandwich Islands is a former name for the
Hawaiian group. The Moriori were inhabitants of the Chatham
Islands, north of New Zealand. Australia (A) refers to the series of
Morant (1927).
Australia. —To test the internal relationship of Australian tribes
43 coefficients have been collated from various sources, and despite
a criticism here and there with regard to technique of measurement
or the number and nature of the traits chosen for comparison, we
can have no doubt about the meaning of the coefficients. The
table indicates that 35 of the 43 coefficients are in the primary
I
•\
s
*
\ <
*
Fig. 25. Baining males, Gazelle Peninsula, New Britain. Front and side views.
255
division of values 0-3, which, as every statistician knows, express a

very close connection between the average measurements for traits
of different groups. Nor can we doubt that the mathematical
relationship between skull measurements signifies great uniformity
of type over the large Australian area of 3,000,000 square miles.
Comparison of photographs, together with the general conclusions
from measurements on living subjects (Howells, 1937a, pp. 24, 25),
confirms this idea of homogeneity. A full discussion of group dif-
ferences based chiefly on the researches of Hrdlicka (1928), Morant
(1927), and Wagner will be found in Wagner's compendium (1937,
pp. 124-127). Somewhat high CRL values appear to justify the
isolation of figures for northern Australia from the rest of the series,
but on the whole we have as good a series for pooling as ever we are
likely to get over so great an area.
Unless quite positive of technique and all other factors entering

into comparisons of two series, I would not stress small differences
of CRL values as a basis for making fine ethnic distinctions between
different parts of Australia. But we know of course that such dif-
ferences exist, and that one or more cranial traits may show a pro-
nounced regional difference.
Australia and Melanesia. — If we turn from the close resemblances

of internal Australian groups to judge the possible relationship of
Australia to other Pacific regions a change in the value and number
of the coefficients will be noted.
A comparison of Australia with Melanesia gives no primary
CRL values, but we have eight secondary values for eight coefficients
that have been worked out between parts of Australia and parts
of New Guinea. Four coefficients (values 3-10) link skull measure-
ments for Australia with series from New Caledonia, and for the
same two regions there are three coefficients of tertiary value avail-
able. For comparison of Australia and the Solomon Islands there are
two coefficients, one secondary, the other tertiary. With a total of
12 coefficients of values 3-10 we can be sure of a fairly close con-
nection between the averages for cranial traits of Australia, parts
of New Guinea and New Caledonia.
Australia and Polynesia. —A shift in the value of the coefficients

is noticeable if we glance down the table of coefficients that link
Australia with Polynesia. We have only four coefficients and these
are all of the tertiary order (10-20). Yet those who have worked
with the CRL method know that to get CRL values of this kind,
there must be a fairly close resemblance between the series of aver-
Ph
257
ages under comparison. We may say that there are undeniable

degrees of likeness between Australian skulls and those of Tahiti
and the Loyalty Islands.
Tasmania. —
The coefficients indicate a relationship between
cranial measurements from Tasmania, Australia, and New Guinea.
There is only one coefficient of the primary order, and this correlates
the measurements of Tasmanian skulls with a very small series from
West Australia. There are, however, five secondary coefficients
linking the Tasmanian measurements with those of West Australia,
New South Wales, and Victoria. The total number of coefficients
linking Tasmania with Australia is 14. Tasmania is linked with
New Guinea by five coefficients, two of them secondary, and three
tertiary. There is a coefficient of the tertiary order showing some
mathematical relationship between skulls of Tasmania and those
of New Britain.
Torres Strait. — Measurements

taken in islands of the strait
between Australia and New Guinea yield coefficients that are mostly
of the secondary order. In all there are 10 coefficients available,
most of which show a fairly close connection between the series of
figures for Torres Strait and southern New Guinea. One coefficient
links Torres Strait with Tasmania, and one indicates relationship
with Australia. There is a coefficient of secondary value linking
Torres Strait with New Caledonia.
Melanesia. —The largest number of coefficients available are
those for internal New Guinea. In order to compare measurements
from various parts of the island 34 coefficients have been worked
out by several different observers. Critics may object to the tech-
nique or to some other matter affecting comparison of certain aver-
ages, but we have here a great consensus of opinion based on work
with the CRL. Apart from the interior tribes, for whom skull
measurements are not available, there is considerable uniformity
in the Papuo-Melanesian types of New Guinea. The results of
investigation by CRL methods have yielded 7 primary coefficients,
23 secondary, and 4 tertiary. The particulars of these are detailed
in the table (pp. 252, 253). In several comparisons (including my
own, p. 127) the craniometrical likeness of north and south coastal
groups is emphasized.
The close resemblance of series of measurements among the
islands of New Caledonia is indicated by 20 primary and 7 secondary
coefficients.
-r
•
lO m
rZ 00
w J-i,
GO
ONNOO
^
£[a>
2goôoooo 00 3 00* X2
pjT-iai — 22 <m ^ t- rl m «o ec ^^
O K
c
>
O
< • cm -<m • i— i •lOOoaiNo^'oooNoom
•
CM «S CO <0 CO Oi i— CO
I CO„£3~ CO -«3< ->3<
, OJ >-H ""tf .-I >- 1—
>
.22
o
2
a. tn
c3 °3
9
o
z "3 o
JS 3
m
ho o o fe 2
pq JX5
x £ pq pq X I O § S £
259
1-1
CM "* t* "*
+
K5 -rj< -n* •n' iD
fC "—'CO ^CO
« 2 c;
as .co
.Qo .t_
A ~ *-<**&•<# CI "& O
.
a
>
Pi z 2 m -* to rf io
H
co
ca
O
H ®o e?> •>* t— 1
3 00
B5
H
fa
fa
^1§2S
<
0) 03
n bo a, 9
-£"2
o
!°
2 ° e
c
•J->
3 2 &
'3
Q •5 csja
Ph ^ 2
"si
O^ ;g£
4-3
03 03 03 03
en
3 >> < 5-s
* s **!§J qa
.-tS-S 03
oj _ S3
51s a I CD
ov&e.
y-PObO
V<
M
03
§ Q >~ S3t3 "2 3b,

OS £ . ss w o
b
P â *-•
b*^- a
5 03
"l
*o n 4i
> SO •*i
Pi
oi a> t- <D ci 3 ° o
|
rt J S 3 03
S £~ „g CO 3
PQ
o PQO b S os if « wO
CD
b P 03 c_i
^2 tr
-
c3
rt^ §
Q
1
a)
tfb..g
H
W a ^ . 03
H £ i 3.-
03 o
w 03 3 03 £ BJ t-i 03
o Pm o C o _1
*" CM 5 J! -
.— Ph bC
CO
2§
o *s Pl,N <=>
^;o5
& s
£ PL,
M— *-< 00
s b <N OS t-
05 • *° -00 • c-
OS
ID
©
o o
u rloo r~\ao —'<x>-~-
rt
s
'3
O ^
M HTfO'^
S3
§
A 2
1~»-
O
O £
w
g «<-H ©
< ^.o
PQ
o
09 GO
H
O
<!
«
«
W S3
- CO ^H
c-j £
'$ Ph O
pq
o
W
o
M T3
S
O
i£ s PQ n o w w a o
261
Linkage between New Guinea and New Britain is definite, also

between New Guinea and the Solomon Islands. There exists also
an undoubtedly close mathematical relationship between series of
skull measurements from New Guinea and New Ireland. All co-
efficients considered, the intra and inter relationships of the Melane-
sian Islands are strong.
—
Melanesia and Polynesia. When proceeding from study of co-
Melanesia only, to those linking Melanesia and
efficients relating to
Polynesia, we note at once a change in the values of the coefficients.
The table shows no coefficients of the first order; but 9 coefficients
of the second order, and 13 of the third order, have been worked
out from the Melanesian and Polynesian series.
A strong mathematical relationship is indicated between measure-
ments from New Caledonia and Easter Island. The latter is in
far eastern Polynesia, but details of measurements indicate that
the craniometry of the Easter Islanders is significantly different
from the general trend of averages for Polynesian islands (von Bonin,
1931). Apparently the Easter Islanders have many craniometrical
features in common with the Melanesians of New Caledonia. Co-
efficients in the section of the table (p. 252) dealing with Australia
indicated a close connection between some Australian series and
those of New Caledonia. The inference seems to be that there is a
mingling of cranial types in New Caledonia. Our table also shows
relationship between craniometrical data from New Guinea-Tahiti,
New Guinea-Loyalty Islands, and between Tahiti-New Britain-
New Caledonia.
Polynesia. — The closeness of internal relationships of Polynesian
islands shown by 21 coefficients of primary value (pp. 252, 281).
is
The Maori crania yield measurements that are very definitely allied
to those of Marquesas Islands, Cook Islands, Society Islands, and
Sandwich Islands. The Maori of New Zealand have also a tertiary
relation with the Moriori, Easter Island, and Loyalty Islands.
Primary coefficients link cranial series of the Marquesas Islands
with the Hervey Islands, Tahiti, Society Islands, Sandwich Islands,
and Moriori. Crania from Easter Island yield measurements showing
a primary CRL value with the Cook Islands and the Loyalty Islands.
If considered as a whole, the 37 coefficients give clear evidence
of the close cranial relationshipswhich exist between islands located
over a vast area of the Pacific Ocean from Hawaii to Easter Island.
The historical and cultural aspects of the problem of racial migrations
in the Pacific Ocean have recently been described by P. H. Buck
(1938). His postulated derivation of Marquesans from Society

Islanders on the ground of place names, genealogies, and cultural
resemblances is in close agreement with the evidence of physical
anthropology. The coefficient of racial likeness between skulls
from these regions is 1.17 ±0.20. Such a low coefficient indicates
that the average measurements for various cranial traits in the two
series are extremely close in value. The cranial data are completely
in agreement with the cultural evidence.
—
Negro Element in the Pacific. Research shows (table, pp. 266-
269) that a series of cranial measurements from Wateita skulls, east
Africa, bears many close resemblances to series of measurements
from Australia, New Guinea, and New Britain. The CRL's given
(tables, pp. 252, 253, 274-281) must not be our only criterion of racial
likeness and unlikeness. We cannot emphasize too strongly the point
that the CRL gives a general comparison of the series, but the co-
efficient does not tell us how many traits have average measurements
so alike that they might be from the same population. Traits so
defined we will refer to as "like traits."
Thetable (pp. 266-269) indicates that in 32 traits compared for

Australia and east Africa 20 of the averages indicate like traits. New
Guinea (Field Museum sample) and east Africa have 12 out of
32 like traits, and the same relationship is found between east
Africa and New Britain. Therefore, the Melanesian samples (New
Guinea and New Britain) have rather more than one-third (12 out
of 32) of Negro traits. We notice that east Africa bears the closest
relation to Australia, then to New Britain.
There is a definite object in including a series of cranial measure-

ments from Old London. We want a control as remote as possible
from the area of our research, just to show how many like traits
occur simply because human skulls have the same phylogenetic
history. We can usually prove some likeness between two series of
skulls, no matter how far they are removed in time and space.
Comparison of Average Measurements

In order to prepare the following tables of averages the whole
of Wagner's (1937) crude data has been recast in the form of weighted
averages. The probable errors for these have been calculated from
the standard deviations for a long Egyptian series (Pearson and
Davin, 1924). The tables include statements to show the number
of samples and the number of skulls on which each average is based.
The need only a few comments, but a student who

figures given
is any of the numerous comparisons which
particularly interested in
are possible can readily apply the formula
Mi- M /V(PE +
2 l )
2
(PE 2 )
2
The term Melanesia needs a word of explanation. Consul-

(2)
tation of Wagner's many
showed that several Melanesian
tables
regions bordering on Polynesia often gave averages that were ob-
viously in good agreement with Polynesian averages, but a poor fit
with averages for Melanesian samples. Melanesia (2) is undoubtedly
geographically Melanesian, but apparently there is considerable
ethnic relationship with Polynesia. Areas included in the heading
Melanesia (2) are from Fiji, New Caledonia, and the Isle of Pines.
Weighted Averages and Their Probable Errors

(Worked Out from the Crude Data of Wagner, 1937)
Region Samples Skulls Average Samples Skulls Average
Cranial Capacity in cc. Maximum Head Length
New Guinea 13 334 1348. 5± 6.21 14 425 183.1±0.28
Australia 6 85 1290.1±12.31 6 90 186.5±0.60
Tasmania 1 33 1264.0±19.77 1 40 182.2±0.61
Melanesia (1) 11 397 1317. 2± ro 11 424 180.8±0.28
Melanesia (2) 7 204 1423. 7 ±
7..95 7 217 188.0±0.39
Polynesia 15 352 1465. 4± 6..05 18 691 185.3±0.22
Field Mus. sample 1 124 1280. 0± 6.87 1 124 177.6±0.35
Maximum Head Breadth Length-Breadth Index

(Cranial Index)
New Guinea 14 425 131.2±0.23 14 423 71.7±0.13
Australia 6 90 130.5=1=0.50 6 90 70.0±0.28
Tasmania 1 60 136.0±0.41 1 43 74.2±0.27
Melanesia (1) 11 423 132.6±0.23 15 694 72.9±0.10
Melanesia (2) 7 217 131.0 ±0.32 7 218 69.7±0.18
Polynesia 18 678 139.5±0.18 18 678 75.4±0.10
Field Mus. sample 1 124 131. 3 ±0.29 1 124 74.0±0.16
Skull Height Basion to Bregma Bizygomatic Breadth

New Guinea 14 388 134.0±0.25 14 360 130.0±0.24
Australia 6 90 133.7±0.53 6 84 135.0±0.50
Tasmania 1 55 130.9±0.46 1 44 131.0±0.49
Melanesia (1) 11 416 135.7±0.25 9 306 132.1±0.26
Melanesia (2) 7 209 140. 6 ±0.35 6 152 136.3±0.37
Polynesia 18 656 139.4±0.20 13 355 135.4±0.24
Field Mus. sample 1 106 131. 6 ±0.33 1 119 128.0±0.29
Upper Facial Height Nasal Breadth

New Guinea 10 273 66.6±0.25 13 363 26.2±0.09
Australia 6 87 69.0±0.44 6 90 27.0±0.19
Tasmania 1 36 62.5±0.47 1 22 26.5±0.38
Melanesia (1) 8 296 66.0 ±0.24 8 300 26.0±0.10
Melanesia (2) 2 111 69.1 ±0.39 6 151 26.5±0.14
Polynesia 12 338 70.7±0.23 14 377 25.9±0.09
Field Mus. sample 1 123 66. 5 ±0.25 1 124 25.3±0.11
Weighted Averages and Their Probable Errors — continued

Region Samples Skulls Average Samples
100 0,R
Orbital Index
0,R
New Guinea 11 332 81.6±0.28
Australia 6 90 76.5±0.53
Tasmania ...
Melanesia (1) 8 401 80.0±0.25
Melanesia (2) 5 144 79. 5 ±0.42
Polynesia 13 340 82.6±0.27
Field Mus. sample 1 123 83. 6 ±0.31
Prosthion P Angle
New Guinea 8 245 78.5±0.21
Australia 6 86 78.0±0.35
Tasmania ...
Melanesia (1) ...
Melanesia (2) 6 282 77. 8 ±0.19
Polynesia 10 255 83.9±0.20
B Angle
New Guinea ...
Australia 6 78 39.2±0.30
Tasmania 1 32 36.9±0.47
Melanesia (1) ...
Melanesia (2) ...
Polynesia 6 183 39.9±0.20
os
O Ol 00 !C O l> t- 0Jt-00a5OC0CMTi<c-T*c-
^ ^COO UJOrH 2rHO Tf ® H
-° ÔOWTfLOCCOSMlOOTfeCWMWOOOSN
°M t-_ <-,
41 41 41
'" -"
H '
-S ""--n
-H
- -H
1
WHHTf ot-oOTfr-lmcMCMcocMCMC^l'^
,M00^'*C]O«D0>01Ci3-<fOOOc0(NrHio
,
°NO t^co^h
0^° —'O
44
v
2^^
Cl-
""#O
'—'t—
i-i
4J
i^-<*0
£it- m
ÔOONt-OONMN^lOt-MCCttrHT-lîn
Ct" i— i-h
con coTI
4-1
co n
4J
co Tl coTl 4J
co Tl
i
'5
"*m cm ^-.io •**

oo
;~1<0<M
Mt-ÔOOXMHlOMOHtOMHOif
"looooo-î-iasoaiMOcoooo-êi^ti^
^HO^CMO^-tfO oo lOOiOONCl^HlOi icM-^'ôooaicMuo
2 -H ^2 ~2 -H -H '2 41 'S-H
C8
Eh
. S^^ w 2
co
w 2'-io "^,o c
T
*eo o c? -
co
co
< )
o->*t-o l^*o
05 10 g^t- oa |-t-
HCOOONOOOHOOHtr-t-iOoOiOWH
co-^t-i—looojcocsit-oiocNiOincoi—icj
'OJOi iTrOt-TflOOt-NtOINOSiocOKKD
"2 41^2 41-2 41^2
,
41
l
2 41
lO OS l-US-ÔTft-COt-OOt-NMiOOMOlCO
x IrHTtHOrHrHTllKlOOOlOtOiaCOrHOOMt-
'MO O o
•
Js PS 2* "* • • • •
•** Irt
•
'COO OOTfOOOO>iONOt-M'
(OOCOIMMlOtCTf
° ^cojj-cojj ^cojj-oojj Icojj-cojj,
'OO^MoSÔTfMO^nOOOWOlOHtDOJONOHtDMHOHCOOO
LtT w ^ C, ^ w ^ C^ w '* -U'- CJ <M
jj JJ - 1
-U 4-1 1 t-i t-i t-i <—i
-f|
i— C- CO OS 00 IOO>»W5t-tOt-M(CMX(MOOaT)iOO
i— rÔO O
rvH..cg..
i
„ ^2tt
ffl^.. c
^..co..
<~.«5t-i CTOJ
a5
i—i t-h i ^«£>
co
i—!CO'-iiXiaiC-0O00CNlC<lt~<X>'-H-^ T-iaj(M ,
OCX! JJ W CN JJ C-CX) JJ W <M JJ -C-CN1 JJ ÔJ JJ i— I

T-l i-( i-( r-l t-H
^H ^sHCO
aiocot-oôo^ô
^jj~ —
t-OOO^ loiomtototcoit-' t-i ^h o a>
^4|C«3 4|-W4|
i
S*JJ
OO
Luo JJ —
KJOIO
!• JJ
lO^T»0»' O i 00' lOCJNr-lrHTfTf. iCOUOOOOt—ic7st~C]iO
'-H
OS lO !OffiOOOOtO<OiBMNM»OOi»MO
^»C0
CM
.— CXI lO ^ O 00
CO •t-i ^srJlT)lO00OO93IOHO00Tj(Oa!m<J100M
CO oo MO CO 00 o >o ^rj'ocsioiO'-HCSicvicsiast-inaicsjcoiuocooo
^2 41
> JJ ÔS JJ T-l T-l T-H T-l CN1
O
00 CM -—-to
CO^C<lT-lCOCX>T-l<lOC<)OOlOt'--IO«Ot-0-*i<
CO©10t-0<XH>tOCOM?OlMt-IMlOOiH
-r- . cm . . as • os • •
^tDO COOOO
-H^.j.-'SOJjvÔJJ— t-JJ
.- jj
OS &0 O cooo Sw jj ~t- 4,
LO t-H OT-tHNaolH/iOiMCTjCOlflOHCOt-N
-t- to £Tco cxi ôwTjucocoHost-OTHTiiTiiooioôOH
loOOWOOOliOOMt-O
w w lOO'* O'^'tcOOlOiOt-t-OMOlOMt-ffit-
'
^ -H ~3 -H « -H 2 41 '2 41"": +i
t- OS OS TfffliOtOtficOTjiHOOOHO'îfloiON
„ CTiM'-o ^- co oo ros «o -OS C- T lO oowooot-coNîoooiot-coajiOHt-
°> sf M n<o
'#oibooo2®0'fTfo^T)iohO)HTjii>o * lO l> T)l Tj |> O Tj Tj (»
m
T) 41
~—
(M 4) IN Tl
jj ^-'t-h jj
co Tl
OO
s
"'ori Tl
JJ 'O JJ
CO Tl
<NI
I
~ co
T3
«OOOON05mt-»iOQOOt-NOP5rtT)i
w <C <£> CO l~>lO <N i*0 HOiNCOMNHl-iOtOOONt-niOOOO
i>
-^
2 --
Zj-Ô JJ
C*- 05
1
K5
° 2°° -^to
o ^3?o
II
"^ o o Jôio f2°oocr>toT-ia5co<Ni-H<-(ico-*oo<M-<*iOTi<
jj Fi-kO
ii r\iO M '-h
'-iH -H"<°-H '-H
4J
n H
^ I
£"CO rH
CJ5
t*
lO
<£> ^ 00 COCO i-H
g-t-H CO £-<NJ iH
CO
^«5
OCOOiOOOt-OOU5HNU500ffioOiHClW
COCOCOtOt-OiOt~E~i--iO-'3<CO?DO"iCOCO
!oOO'*oOO'*OOCOIN <*«>WON'ÔJt> ,
!.?£> JJ CM i-H i-li-l
* *
q-o
CO 00 00 O Ô! CO ;T^ OOO00Ot-Oi0t-«3t»NClt-HH(0
~ t-h ,~v<o co Cî-h rn ,~.t- co
Cv ^ w5
ojusoqcoio-îoutKMOst-cocciot-ooo
i-i
CO
iJ Nô iooo 2^0 ^mo J'ho ^lOOOtcooWHt-'iiMooooajiOTfwoci
. . £t0 • • . . . . . ,
* # # *
(M
^00 CO ^00
5 -^ 10
£pO HTfWWo(0«0!CO0)C0501OC0t-C0t>
H
<x> crs o o o 00 00 ai as 00 co •>* t- t-
^5 "^
£
•«
°MOÔio2oOCOOO
"CO — CO ~ca "~^<m
JJ "
41 11 11
00O 2 00u, I-H
^' c>0 —iW-^hOtJioÔOOco*
CO JJ C!-CO JJ I-H I-H
'
iH i-H
41 -h
îo 1-1
«0 «D »0
Oco 1-1 ^00 co
-"tf Tf Ci^tÔHOOt-tDOWOOffiOMHCO
—* 2
°MO
^-Ô co
••••••••••••••»..
i^co •* r^-00 r-ioot-'Ot-cotDLOcooicocoooiocooo
^CO 2^0
* "
fVl " • • *
CJi • * • • •
£
"~
t«(00
£j-C0 -U C-CO
CO-Ô
jj 4)
*— £^ -** O ^t0005IM' l HCOOO'*OWiJ(NH«)iOTji
, ,
ti w
* * * *
CO CO -# CO CX> t- C- tD "# CD lO lO CO CO Tl< -<J( 00 Oi
^^••"-l ••00- -SO ••<£>• -OS --ô coooaicoi-Hcocoôcrsocot-icco-î-ico _ _ 1-1
rj ;~ t- O Mcoo OJ 1—IO CO t- O IOOO 5CCDOHOOifiJiHOOiNC3MXNCOClX

C-** "* "-"tO
-J-|
w "* w >0 "- JJ J-| i-Hi-H JJ JJ "f JJ i-H i-H i-H i-l i-H
00C0t>Nif00«!lOtDOt>^?Ct-(Ol0!£l
o
I
to
CO i-H ^H CO CTco
-»~-C0 COCOOt-COcO-^t-OOO-^t-CO-'^COi-io
-—-«* i-H
< WOO w
OO COOio *-'"* COCOO ™HO lOOJOCOTîOlOCOCOiO-^tiO-^COCOCOOO-^
^ d-^f '-'CO
"~~<
JJ rH -W 1
41 JJ 44
41 -H
>
-"*•** ^-vOCO i-H

co
CO ^"i-H CO
o ^t- co o 00 as t- vo 00 o co 00 o o 10 10 o 00
-**
COOOiOi-iO'îiOCOOCOlOi-ICOiOO'*
„«o ••«?.-. 00 ••î. ••»..£-.
O aicooNHoCifflOMHO^wowaoTfowaHHOOMi*
^W JJÎO ll^'Cl JJ~— IC .-IO co 1-1 co co 1-1
'H
II
+1 -H ^ i-H i—l i-H 1—1
-=1 CO
s
.5 c
c3 O
-4^-73
M ^ PQ <?:
i-H t-.
"o3
o
5
S2Co
»-( to oo CO
os
< WO>0000OC0tfi'<*t0CCW0000i0M
i—
k; i-h
w V* CM
oo
-<M CO OiO-*J<CO-'#COCMOO-r} -'tfCJSOOCOCOCM"'3'
,
Z^
-^ wt
-H ^
3wON050 0000M«30r1'^0®'<fO
w <° ~*- w <°
-H
"
-H -H -H -H
00 t- <* lO CO CM lO •
lOSOOCM^fOCOOSCl
I HHM <M H
-f
^<x> co /î-h ,—.0 co f^CM
(M t-t-MOlOHt-OOO^t-NHO^H
^m co COCOCO-^t^COOOOOCOiOOlCO-'tftÔSOS
i't- i-h t-i i-i
HN 5«ONfflOOOOlOMOOr;oOOtCIMO NHMNOHOMtOIMTjiasiniONTt
~°°-H -H -H M
CO ">* to T)lt-H[-t-!ClN(OOOOMr- I CM t- C- CO
CM ** ,~.iO CXI -~-0 CO J-rH t-i ^CO CO CMt-00iO^<eOOSt-OJt-00a0iO-*d<t-CO
CO
«<N O ooonhooooowô
t-
• <M• " • •
cos£>cococor-4->3<oocot~o?r>t~T-Ht~oo
"CO CO ^~.00 CO
OOHTfO<0«Dt-WO>0(DO»HN
OOOl-MNOOOOHHt-iONt-iNt-H
— — < >
"3 V*1
S oÔMHoSÔMooo^NoSno
"oo
.
—
•
"oo —
m
CJ5
'
•
oo _u
• '
41
' OS • •
'oo _u iCioo
*"'*'
•
II CJ-oo ij
COOi0^t"0]COt-i—INU3TfNmt-Hî
z
Two distinguishing features of the Australian face are a low orbital

index, which indicates rectangular eyesockets, and the highest nasal
index, with the exception of that for the Tasmanians. The face is
relatively prognathous.
The Tasmanian skull has the smallest cranial capacity of the
whole series, has the lowest vault. The cranial index of 74.2 ±
and it
0.27 is close to our Field Museum New Guinea sample with an index
of 74.0 ±0.16. The Tasmanian skull has the lowest facial height
(62.5 ±0.47 mm.) for the whole series. This trait gave the typical
scowling appearance of Tasmanians. The same facial expression is
noticeable in Bushmen of South Africa, whose upper facial height
(62.1 ±0.51 mm.) is a little lessthan that of the Tasmanians (Kitson,
1931, p. 292). The nasal index for Tasmanian skulls is the highest
of the whole series. Study of facial angles shows that Tasmanians
have the smallest B angle in the series.
A comparison of a Field Museum New Guinea sample with
averages of 13 samples comprising 334 skulls from that island shows
resemblances and differences (tables, pp. 259-261; 264, 265). The
Museum sample shows a shorter skull with lower cranial capacity
than that for Melanesia (1). But we must bear in mind that the
Field Museum sample may be regarded as typically Papuan-Me-
lanesian, whereas all samples, and especially Melanesia (2), may
have admixture of both Australoid and Polynesian traits.
Conclusions
The main conclusions are:
That the CRL a valuable means of making comparisons
is
between two average measurements for different traits.

series of
But we must have regard for the number of traits which have no
significant difference. The CRL will give only a general measure
of likeness and unlikeness, and (like any other average) the CRL
obscures the number of identities; extremely close likenesses may
be obscured by disparities between pairs of traits.
Unless we are working with ideal data that give exactly what
we want in size of samples, traits, and technique of measurement, we
must not stress too much the small differences of CRL values.
We have a right to say, however, that CRL values from 0-3 express
a very close mathematical relationship between two series of aver-
ages, and that CRL values from 3-10 leave us free to think that, in
a statistical sense at least, the averages for the compared series are
closely related. We have then to judge whether on ethnological
principles the mathematical relationship implies ethnic affinity.
Consistent results in CRL values can sometimes be obtained with

small samples of homogeneous material. This is apparent from
consideration of Wagner's (1937, pp. 123, 165) coefficients expressing
relationship between cranial series within Australia and within New
Caledonia. The numerous small samples yield many coefficients
between and 10 in value, and these coefficients are consistent in
showing homogeneity of cranial measurements. But we must not
infer from these instances that small samples could always be used
with reliability; much depends on the homogeneity of the individ-
ual skulls which make up the total series.
Comparison of craniometrical data from east Africa, Australia,
New Guinea, and New Britain suggested a considerable mixture
of Negro traits in the Pacific regions. We made an extensive survey
of peoples of Australia, Melanesia, and Polynesia by (1) analysis
and grouping of 244 coefficients of racial likeness; (2) comparison
of averages, modes, and ranges of measurements. The smallest
coefficients are those linking craniometrical measurements within
Australia, within New Guinea, among the islands of New Caledonia,
and among the Polynesian islands, even though they are far apart.
There are mathematical grounds for believing in the probability
that Tasmanians are an Australoid-Melanesian people. Values of
the CRL are in harmony with general views on the inter-relationship
of Melanesian peoples, and the mixture of Polynesian, Australoid,
and Melanesian cranial characters.
The correlation of craniometrical data with the results of blood
tests on various racial groups offers an interesting field for future
research.
Finally, physical anthropologists should co-operate internation-
more uniformity in technique of measurement. It is
ally to secure
fundamental to know what we intend to measure, how we shall
make the measurements, and what we are aiming at in compara-
tive study.
At present, tables of measurements are published in so many
different forms that much time is lost in extracting data. The form
adopted in "Biometrika" is one that favors rapid comparative study.
Samples which are available for comparative study are often
unsatisfactory both in quality and quantity, and a research worker
is between Scylla and Charybdis; he wishes to avoid the use of
small samples, but fears to pool measurements that are not sufficiently
homogeneous. A critical survey of measurements should (1) elimi-
nate skulls of uncertain provenance; (2) cast out measurements
made by samples that

different or unreliable techniques; (3) reject
are doubtfully sexed; (4) on having for measurement an
insist
adequate number of those traits which are of the greatest impor-
tance in showing cranial differences; (5) retain the chosen traits for
all comparisons.
All these requirements are scarcely ever met by samples we
wish to compare, and a severely critical attitude results in so
much elimination of data that insufficient material remains for
comparative study.
Finally, I must take full responsibility for all the work in this
monograph. And with regard to the last chapter, in which some
comparative study is attempted, I am definitely between two schools.
On the one hand, there are those who believe that much more might
be squeezed from the data. For example, by "reducing" all co-
efficients, we might by mathematical manipulation accept the cheer-
ful belief that we have 100 skulls in each sample. Furthermore, it
might have been possible to attempt further adjustments for
known differences in technique, and so agree to the inclusion of
data that have been disregarded. Some physical anthropologists
are willing to go farther than others in making "adjustments" and
"allowances."
On who would condemn the use
the other hand, there are critics
of the coefficient entirely, and for this reason I have favored what
seems to be a middle course, which results in the suggestion of some
tentative conclusions of a general nature.
APPENDIX
Tables Giving Coefficients of Racial Likeness
The is an acknowledgment of the source.
letter following a coefficient B = Bonin,
G. von (unpublished data including some unpublished figures from New Ireland,
supplied to Dr. von Bonin by Dr. 0. Schlaginhaufen, see p. 278); Ha. = Hambly,
W. D.; Hr— W=data of Hrdlicka, as treated by Wagner, K. A.; M=Morant,
G. M.; W= Wagner, K. A. For detailed analysis of these tables consult tables on
page 252, and text (pp. 253-263).
273
a >»
T3
<
o
W fa
S 3
03
£ £
55 3
o
<
W
P
o
Q
<
>> co
03 C
q 5
BIBLIOGRAPHY
Bauer, L.
1915. Beitrage zur Kraniologie der Baining (Neu-Pommern). Archivio per
l'Antropologia e la Etnologia, Bd. 14, pp. 145-193.
Benington, R. C, and Pearson, K.
1911. Cranial type-contours. Biometrika, vol. 8, pp. 123-201.
Berry, R. and Robertson, A. W. D.
J. A.,
1914. The place in nature of the Tasmanian aboriginal as deduced from a
study of his calvaria. His relation to the Australian aboriginal. Proceedings
of the Royal Society of Edinburgh, vols. 34-35, pp. 144-189.
and Cross, K. St.

1910. A
biometrical study of the relative degree of purity of race of the Tas-
manian, Australian, and Papuan. Proceedings of the Royal Society of Edin-
burgh, vol. 31, pp. 17-40.
Bijlmer, H. J. T.
1923. Anthropological results of the Dutch scientific Central New-Guinea
expedition, 1920, followed by an essay on the anthropology of the Papuans.
Nova Guinea, vol. 7, pp. 355-448. Leiden. See extensive bibliography (241
titles quoted), pp. 442-448.
1939. Tapiro Pygmies and Pania Mountain Papuans. Nova Guinea, newser.,
vol. 3, pp. 113-184.
BONDY-HOROWITZ, E. VON
1930. Beitrage zur Anthropologie von Nordost-Neu-Guinea. Bibliography,
pp. 148-152. Vienna.
Bonin, G. VON
1931. A contribution to the craniology of the Easter Islanders. Biometrika,
vol. 23, pp. 249-270.
1934. On the size of man's brain as indicated by skull capacity. Journal of
Comparative Neurology, vol. 59, pp. 1-28.
1936. On the craniology of Oceania. Crania from New Britain. Biometrika,
vol. 28, pp. 123-147.
and Morant, G. M.
1938. Indian races in the United States. A survey of previously published
cranial measurements. Biometrika, vol. 30, pp. 94-129.
Broek, A. J. P. van den
1913. Uber Pygmaen in Niederlandisch-Siid-Neu-Guinea. Zeitschrift fur Eth-
nologie, vol. 45, pp. 23-44.
1923a. Untersuchungen an Schadeln aus Niederlandisch-Sud-West-Neu-
Guinea. Nova Guinea, vol. 7, pp. 163-233. Leiden.
1923b. Zur Anthropologie des Bergstammes Pesechem, im Inneren von Nieder-
landisch-Neu-Guinea. Nova Guinea, vol. 7, pp. 233-276.
Buchner, L. W. G.
1914. A study of the curvatures of the Tasmanian aboriginal cranium. Pro-
ceedings of the Royal Society of Edinburgh, vols. 34-35, pp. 128-143.
Buck, P. H.
1938. Vikings of the sunrise (a story of Pacific migrations). New York.
Buxton, L. H. D., and Morant, G. M.
1933. The essential craniological technique. Journal of the Royal Anthro-
pological Institute, vol. 63, pp. 19-47.
282
Bibliography 283
Cave, A. J. E.
1936. On the human crania from New Guinea, collected by Lord Moyne's
expedition, and now in the Royal College of Surgeons, London. Windmill
Press, Kingswood, Surrey. Privately printed.
Davies, A.
1932. A re-survey of the morphology of the nose in relation to climate. Journal
of the Royal Anthropological Institute, vol. 62, pp. 337-359.
De Beaufort, L. F., Pulle, A. A., and Rutten, L.

1926. Nova Guinea. Resultats des expeditions scientifiques a la Nouvelle
Guinea, vol. 15. Leiden. See P. Wirz.
Dorsey, G. A.
1897. Observations on a collection of Papuan crania. Field Columbian Mu-
seum Publication 21, Anthr. Ser., vol. 2, No. 1.
Encyclopaedia Britannica
1929. General account of the history, geography, and ethnology of New Guinea.
14th ed., vol. 16.
Fisher, R. A.
1934. Statistical methods for research workers. Edinburgh.
1936. Coefficient of racial likeness and the future of craniometry. Journal of
the Royal Anthropological Institute, vol. 66, pp. 57-63.
Graf, L. A.
1931. Uber eine Schadelserie aus West-Neu-Guinea. Universitat, Zurich.
Gray, H. (and collaborators)
1927. Auricular head height technique (five articles). American Journal of
Physical Anthropology, vol. 11, pp. 301-351.
Gray, J.
1901. Measurements of Papuan skulls. Journal of the Royal Anthropological

Institute, vol. 31, pp. 261-264.
Haddon, A. C.
1915-16. Note antropologiche sui Papua occidentale della Nuova Guinea
Inglese. Rivista di Antropologia, vol. 20, pp. 3-20.
Hambly, W. D.
1926. Origins of education among primitive people. Vital statistics, pp. 22-27.
London.
1937. Source book for African anthropology. Field Museum of Natural
History, Anthr. Ser., vol. 26, pp. 161-220.
1938. Anthropometry of the Ovimbundu. Field Museum of Natural History,
Anthr. Ser., vol. 25, No. 2. Vital statistics, pp. 65-71.
Hooke, Beatrice G. E.
1926. A third study of the English skull with special reference to the Farringdon
street crania. Biometrika, vol. 18, pp. 1-55.
Howells, W. W.
1933. Anthropometry and blood types in Fiji and the Solomon Islands. Ameri-
can Museum of Natural History, Anthropological Papers, vol. 33, pt. 4,
pp. 279-339.
1937a. Anthropometry of the natives of Arnhem Land and the Australian race
problem. Papers, Peabody Museum of American Archaeology and Ethnology,
Harvard University, vol. 16, No. 1. Data collected by W. L. Warner, pp.
1-97.
1937b. The designation of the principal anthropometric landmarks on the
head and skull. American Journal of Physical Anthropology, vol. 22, pp.
477-494.
Hrdlicka, A.
1920. Anthropometry. Philadelphia.
1928. Catalogue of human crania in the United States National Museum
collections. Proceedings of the United States National Museum, vol. 71.
New Guinea, pp. 92-95. Washington.
Isserlis, L.
1914. Formulae for determination of the capacity of the Negro skull from
external measurements. Biometrika, vol. 10, pp. 188-192.
Kitson, E.
1931. A study of the Negro skull with special reference to the crania from
Kenya Colony. Biometrika, vol. 23, pp. 271-314.
Krzywicki, L.
1935. Primitive society and its vital statistics. London.
Lewenz, M. and Pearson, K.
A.,
1904. On the measurement of internal capacity of cranial circumferences.
Biometrika, vol. 3, pp. 366-392.
Lewis, A. B.
1932. Ethnology of Melanesia. Field Museum of Natural History, Anthro-
pology Guide, part 5.
Luschan, F. VON
1907. Sammlung Baessler. Schadel von Polynesischen Inseln. Veroffent-
lichungen aus dem Konigl. Museum fur Volkerkunde in Berlin, vol. 12.
Macdonell, W. R.
1904. A study of the variation and correlation of the human skull with special
reference to the English crania. Biometrika, vol. 3, pp. 191-244.
Mantegazza, P.
1877. Studii antropologici ed etnografici sulla Nuova Guinea. Archivio per
l'Antropologia e la Etnologia, vol. 7, pp. 137-172.
and Regalia, E.
1881. Nuovi studi craniologici sulla Nuova Guinea. Archivio per l'Antro-
pologia e la Etnologia, vol. 11, pp. 149-185.
Martin, R.
1928. Lehrbuch der Anthropologic, 2nd ed., 3 vols. Jena.
Meyer, A. B.
1877. Uber 135 Papuaschadel von Neu-Guinea und der Insel Mysore. Mit-
teilungen des Konigl. Zoologischen Museums zu Dresden, Heft 2, pp. 61-100.
1886. Masse von Schadeln aus dem ostlichen Teil des ostindischen Archipels.
Verhandlungen der Berliner Gesellschaft fur Anthropologie, pp. 319-321.
MORANT, G. M.
1923. A first study of the Tibetan skull. Biometrika, vol. 14, contour drawings,
pp. 227-246.
1927. A study of the Australian and Tasmanian skulls. Biometrika, vol. 19,
pp. 417-440. Cf. measurements with those given by Hrdlicka, 1928.
1928. A preliminary classification of European races based on cranial measure-
ments. Biometrika, vol. 20B, pp. 300-365.
1931. A study of the recently excavated Spitalfields crania. Biometrika, vol.
23. Reduced coefficient of racial likeness, p. 214.
, Tildesley, M. L., and Buxton, L. H. D.

1932. Standardization of the technique of physical anthropology. Man, No.
193.
Bibliography 285
Pearson, K.
1926. On the coefficient of racial likeness. Biometrika, vol. 18, pp. 105-117.
1928. Note on standardization of method of using the coefficient of racial like-
ness. Biometrika, vol. 20B, pp. 376-382.
and Davin, A. G.
1924. On the biometric constants of the human skull. Biometrika, vol. 16,
pp. 328-363.
Pitt-Rivers, G. H. L. F.
1927. The clash of culture and the contact of races. London.
Sarasin, F.
1916-22. Anthropologie der Neu-Caledonier und Loyalty-Insulaner. Berlin.
Saunders, C.
1922. The population problem. London.
SCHLAGINHAUFEN, 0.
1912-14. Anthropometrische Untersuchungen an Eingeborenen in Deutsch-
Neuguinea. Abhandlungen und Berichte des Konigl. Zoologischen und An-
thropologisch-Ethnographischen Museums zu Dresden, vol. 14, No. 5, pp. 5-
82. With a bibliography of 50 titles.
1915-16. L'indice cefalico in alcuni gruppi umani della costa settentrionale
della Nuova Guinea. Rivista di Antropologia, vol. 20, pp. 3-12.
1929. Zur Anthropologie der Mikronesischen Inselgruppe Kapingamarangi
(Greenwich Islands). Archiv der Julius Klaus-Stiftung fiir Vererbungsfor-
schung, Sozialanthropologie und Rassenhygiene. Band 4, Heft 3-4, pp.
220-287.
Scott, J. H.
1894. Contribution to the osteology of the aborigines of New Zealand and of
the Chatham Islands. Transactions of the New Zealand Institute, vol. 26,
pp. 1-64.
Seligman, C. G.
1909. A classification of the natives of British New-Guinea. Journal of the
Royal Anthropological Institute, vol. 39, pp. 246-274, with plates and
description of living subjects.
1910. The Melanesians of British New Guinea. Cambridge, England.
Seltzer, C. C.
1937. A critique of the coefficient of racial likeness. American Journal of
Physical Anthropology, vol. 23, pp. 101-109.
Speiser, F.
1923. Anthropologische Messungen aus den Santa Cruz Inseln. Archiv fiir
Anthropologie, N. F. 19, pp. 89-146.
Stewart, T. D.
1937. An examination of the Breitinger method of cranial capacity determina-
tion. American Journal of Physical Anthropology, vol. 23, pp. 111-121.
Stoessiger, B. N., and Morant, G. M.

1932. A study of the crania in the vaulted ambulatory of Saint Leonard's
church, Hythe. Biometrika, vol. 24, pp. 135-202.
Thomas, O.
1884. Account of a collection of human skulls from Torres Straits. Journal
of the Anthropological Institute, vol. 14, pp. 328-343.
Thomson, A., and Buxton, L. H. D.

1923. Man's nasal index in relation to certain climatic conditions. Journal
of the Royal Anthropological Institute, vol. 53, pp. 92-122.
TlLDESLEY, M. L.
1921a. Table for calculating occipital index. Biometrika, vol. 13, pp. 261-
262.
1921b. A first study of the Burmese skull. Biometrika, vol. 13, contours, pp.
185-229.
1927. Determination of the cranial capacity of the Negro from measurements
of the skull or the living head. Biometrika, vol. 19, pp. 204-206.
1938. Measurements of head length and breadth. The systems of Broca,
Topinard, Monaco agreement, British Association, Martin, compared. Man,
Nos. 103, 144.
1939. Congres International des Sciences Anthropologiques et Ethnologiques:
Comite de Standardisation de la Technique Anthropologique. Man, No. 68.
Todd, T. W.
1923. Cranial capacity and linear dimensions in White and Negro. American
Journal of Physical Anthropology, vol. 6, No. 2, pp. 98-194.
and Lyon, D. W., Jr.
1924. Endocranial suture closure: Its progress and age relationship. American
Journal of Physical Anthropology, vol. 7, pp. 325-383.
Turner, W.
1884. Report on the scientific results of the voyage of H. M. S. Challenger,
1873-1876. Report on the human skeletons and the crania. Zoology, vol. 10,
pt. 29, p. 96; vol. 16, pp. 4-128.
1909. The craniology, racial affinities, and descent of the aborigines of Tasmania.
Transactions of the Royal Society of Edinburgh, vol. 46, pp. 365-404.
1910. The aborigines of Tasmania. Transactions of the Royal Society of Edin-
burgh, vol. 47, pp. 411-454.
Wagner, K. A.
1937. The craniology of the oceanic races. Skrifter Utgitt av Det Norske
Videnskaps Akademi i Oslo I. Matematisk-Naturvidenskapelig Klasse.
No. 2, pp. 7-193. With bibliography of about 150 titles.
Weisbach, M.
1890. Der Maorischadel. Mitteilungen der anthropologischen Gesellschaft,
vol. 20, pp. 32-37. Vienna.
Wilder, H. H.
1920. A laboratory manual of anthropometry. Philadelphia.
Wirz, P.
1926. Untersuchungen an Schadeln und Skeletteilen aus dem Gebiet der
Humboldt-Bai und dem siidlichen Kiistengebiet von Hollandisch Neu-Guinea.
Nova Guinea, vol. 16, pp. 149-250. Leiden.
Wollaston, A. F. R.
1912. Pygmies and Papuans. Measurements of Tapiro Pygmies, Appendix B,
p. 321. New York.
Wood-Jones, F.
1936. Skulls from the Purari Plateau, New Guinea. Journal of Anatomy,
vol. 70, pt. 2, pp. 405-409.
WUNDERLY, J.
1939. The cranial and other skeletal remains of Tasmanians in collections
in the Commonwealth of Australia. Biometrika, vol. 30, pp. 305-340.
Yule, Udny G.
1937. An introduction to the theory of statistics. London.
INDEX
Abbreviations, for technical terms of Cranial capacities, differences of, ac-
measurements, 91 cording to methods of measuring,
Africa, coefficients of racial likeness 259; frequency distribution of, for
with New Britain, 270; Negro skulls New Guinea, curve of, 226; technique
compared with samples from New and remeasurement of 50 skulls, 92
Britain, New Guinea and Australia, Cranial indices, New Guinea, curve for,
252; skull contours of the Wateita 29; see all Tables of measurements
tribe, 239
Age, of skulls in the collection, 122-125 Dallmannhafen, ancestral skulls from,
Aims of research, 88 87
Alveolar prognathism, 207 Data, and their defects, 251, 271, 272;
Ancestral skulls, 87 see Tables
Angles of the face, denned, 91 Death rates, estimated from age of
Appendix, tables giving coefficients of skulls in collection, 122-125
racial likeness, 274-284
Definitions of measurements, 91
Arcs, of New Guinea crania, see Tables Demography, in relation to age and
of individual measurements and sex of skulls, 124, 125
curves, 231-234 Diagrams, curves of frequency distribu-
Asterion, wormian bones at, see Tables tions for skulls of New Guinea, 226-
of individual measurements, 130-216
238; of skull contours for New
Australia, average cranial measure- Guinea and other regions, 239-243;
ments, comparative study and tabu- see also Frequency distribution
lation of, 264-269; coefficient of
racial likeness for, 253; internal and
external racial relationships, 256, Easter Island, coefficients of racial like-
257; intra-coefficients of racial likeness, 252, 253, 262, 280, 281
ness, 274; skulls compared with those Egyptian skulls, standard deviations
of other Pacific regions, 252; skull for, used in estimating probable errors
contours, 239 of other series, 128, 129, 221

Averages, of cranial measurements for Endocranial sutures, used for estimat-
various Pacific regions compared in ing age of skull, 122
detail, tables, 264-269; for measure- Epipteric bones, 195, 201
ments of individual skulls of New Errors of measurement, see Technique,
Guinea, 214-221 92
Ethnic affinity, and coefficient of racial
likeness, 270
Baining tribe, Gazelle Peninsula, photo-
graphs, 255
Bamu River, skulls from, 89 Facial angles, accuracy in remeasure-
Bizygomatic widths, New Guinea, ment of, 96
curve for, 235 Facial indices, New Guinea, frequency
Borbor, ancestral skulls from, 87 curve for, 237
Breadth of skulls, New Guinea, curve Facial triangles, New Guinea, diagram
for, 228 showing sex differences of, 238
Bure, ancestral skulls from, 87 Family groups, material not sufficient
for study of, 87
Fly River, skulls from, 89
Coefficient of racial likeness, 244-245;
Formulae, for approximate standard
comparative study of skulls from
deviations, 128, 129; for coefficient
various Pacific regions, 264-269; of
of racial likeness, 127, 128; for com-
skulls from northern and southern
paring differences between two aver-
New Guinea, 127; value of, discussed, ages, 264; for measuring cranial ca-
270
pacity, 94, 95; and pooling of data,
Comite de Standardisation de la Tech-
126
nique Anthropologique, 90 New
Frequency distributions, traits of
Conclusions, 270-272
Guinea skulls plotted, 222-238
Contours, of skulls from New Guinea
and other regions, 239-243
Cook Islands, coefficients of racial like- Gazelle Peninsula, New Britain, tribes
ness, 252, 253, 281 of, illustrated, 255, 257
287
Geography of New Guinea, 87; place 244-281; inter-coefficients of racial
names denned, 256; village groups, 88 likeness for, 278-280
Girara, head-hunters of, 87 Merauke, head-hunters of, 87; skulls
from, 89; skulls measured at, 87, 225
Hawaii, see Sandwich Islands Metopic suture, 153, 207
Head-hunters, of Girara, 87; of Me- Migrations, of Pacific peoples, 253, 256,
rauke, 87; of Purari Delta, 87; of 263
Strickland River, 87; and variability Modal values, as shown in frequency
of skulls in the collection, 223 distributions, 222-238
Heights of skulls, New Guinea, curve Mortality judged from age of skulls,
for, 230 122-125
Hienghene, see New Caledonia, 277 Murik, ancestral skulls from, 87
Hythe skulls, standard deviation for Mustard seed, for measuring skull
series, used in estimating probable capacity, 92; see Skull capacity and
errors for other series, 128, 129, 221 technique
Indices, curves of frequency distribu- Nasal indices, New Guinea, curve for,
tion for, 226-238; list and definitions 236
of, 91; skulls of New Guinea, curves
Negroes and Melanesians, likeness of
for, 229-237 traits, 253, 263
Individual measurements, tables for
Negro skulls, compared with those of
skulls of New Guinea, 130-213
Pacific regions, 263-269; contour
Interorbital width, difficulty of measur-
drawings of, 239-243; formula for
ing accurately, 96, 97
measuring capacity of, 95
New Britain, coefficients of racial like-
Juvenile skulls, eruption of teeth in, 223; ness for, 252, 278; skull contours of,
not measured, 122 239; tribes of, illustrated, 255, 257
New Caledonia, coefficients of racial
Kanala, see New Caledonia, 277 likeness for, 252, 278, 280; intra-co-
Kirau, ancestral skulls from, 87 efficients of racial likeness for, 177;
Kone, see New Caledonia, 277 measurements of, compared with
those of New Guinea, 262
Lambda, wormian bones at, see Tables New Guinea, average cranial measure-
of individual measurements, 130-216 ments, comparative study and tabu-
Length of skulls, New Guinea, curve lation of, 264-269; coefficients of
for, 227 racial likeness compared with those
Lifou, see New Caledonia, 277 of New Ireland, Solomon Islands,
Local groups, measurements classified Torres Strait, New Caledonia, and
for, 87; of New Guinea, average New Britain, 278; infra-coefficients
measurements for, 214-221; of New of racial likeness for, 275; summary
Guinea, measurements for, by dif- of tables of measurements from, given
ferent observers, 259-261; pooling by various anthropologists, 259-261;
of measurements for, 126-129; tabu- see Hambly's Table of individual
lation of crania from, 88 measurements, 130-221
Localities, see Tables of individual New Ireland, coefficients of racial like-
measurements, 130-213 ness, 262, 278
London (Old) crania, used in compara- New South Wales, see Australia
tive study with crania of Pacific
New Zealand, see Maori, Moriori, and
regions, 266-269
Polynesia
Loyalty Islands, coefficients of racial
Northern Territory, see Australia
likeness for, 252
Occipital torus, notes on, see Tables,

Maori, coefficients of racial likeness,
130-213
252, 253, 281
Mare, New Caledonia, 277 Ouvea, 277; see New Caledonia
Marquesas Islands, coefficients of racial
likeness compared with those of other Palate, two instruments for measuring
Pacific regions, 252, 281 depth of, 96
Melanesia, averages for different re- Papua, see Melanesia, New Guinea
gions of, 264-269; coefficients of Place names, in New Guinea, 214, 221,
racial likeness for, 252, 253, 258, 262; 261; tabulated, 88, 89; see Tables of
comparative study of skull traits in, individual measurements, 130-213
Index 289
Polynesia, average cranial measure- Table, of abbreviations used for meas-

ments, comparative study and tabu- urements, 91; of alpha values, ob-
lation of, 264-269; coefficients of tained in working out coefficient of
racial likeness of, discussed, 252, 253, racial likeness, 247, 248; averages of
262; coefficients of racial likeness of, alpha values obtained in 15 group-
compared with coefficients of Melan- ings, 250; average measurements for
esian regions, 279; inter-coefficients local groups, 214-221; coefficient of
of racial likeness for, 281; skulls of, racial likeness, 274-281 coefficient of
;
compared with those of Australia, 257 racial likeness, skulls of New

Guinea,
Pooling of data, 126-129; for Australian 246; coefficient of racial likeness for
tribes, 257; difficulty of, 271; and Pacific area, classified according to
sex differences, 225 specific area, 252, 253; confusion of
Probable errors, in relation to the com- tabulations in research work, 271;
parison of averages for different of cranial capacities calculated and
skulls, 264-269; considered in relation measured, 94, 95; of craniometric
to their respective averages, 264; averages for skulls of New
Guinea,
short way of calculating, 128, 129, 221 results obtained by different observ-
Prognathism, alveolar, 207 ers, 259-261; of experimental test of
Pronunciation of place names, 88 variation of coefficient of racial like-
Purari Delta, head-hunters of, 87 ness with nature and number of
traits, 250; of individual measure-
Queensland, see Australia ments for skulls of New Guinea, 130-
213; of measurements of skulls taken
in situ near Merauke, 223; of palatal
Racial likeness, north and south New
depth and interorbital width, first
Guinea, 127; see Coefficients of racial
and second measurement, 96, 97; of
likeness; see also under place and
tribal names, e.g. Melanesia, Austra-
place names in New Guinea, 88, 89;
of remeasurement of fifty male skulls
lia, Maori
Remeasurements, accuracy of, 92-97; from northern New Guinea, 98-121;
of sex ratios, 224; of skull contour
tables of remeasurements, 98-121
measurements, 240-242
Tahiti, coefficients of racial likeness for,
Sampling, 271, 272; see Difficulties,
252; coefficients of racial likeness
Technique with other Pacific regions, 281
Sandwich Islands, coefficients of racial
Tasmania, average cranial measure-
likeness with other Pacific regions,
ments, comparative study and tabu-
252, 281
lation of, 264-269; coefficient of cor-
Sepik River, skulls from, 88
relation for, 253; coefficient of racial
Sex, and age in the collection, 130-213; likeness for, 275; skulls of, compared
characters of, 87; differences of in with those of Australia and New
New Guinea collection, 222, 238; and Guinea, 258; skulls of, traits dis-
mortality rates, 121-125; ratios of,
cussed, 270
for measurements from New Guinea
Technical terms used, 91; see headings
and other regions, 224
of columns of tables
Shot, used for measuring skull capaci-
Technique, co-operation among physical
ties, 93
anthropologists needed, 271 of meas-
;
Sissano, ancestral skulls from, 87
urements, 90-129; and need for uni-
Society Islands, coefficients of racial
formity in making measurements, 257
likeness for, 252
Teeth, eruption of, in juvenile skulls,
Solomon Islands, coefficients of racial
223; see occasional notes in Tables of
likeness for, 252, 278; and New
individual measurements, 130-213
Guinea measurements compared, 262
Standard deviations, of skull measure-
Topography of village groups, 88, 89
ments of northern and southern New Torres Strait, coefficient of racial like-
Guinea, 128; see Probable errors and ness for, 253, 278
Tables Traits of the skull, and their relative
Statistical difficulty, 244; see Coefficient values for estimating racial differ-
of racial likeness ences, 250; see Tables of alpha values,
Strickland River, head-hunters of, 87 247, 248
Styloid processes, 201
Sutures, and age of skulls, 122; peculiar- Universalstereograph, for drawing skull
ities of, 153; see Wormian bones contours, 243
Uraniscometer, for measuring depth of Wateita (tribe of east Africa) skulls
palate, 96 compared with other series from the
„. , . . ,. Pacific, 239, 263-269
Victoria, see Australia nr .
Water a *
used in
for measuring skull capaci-
Vital statistics, judged from ages of
ties94
skulls, 121-125 w
Wormian
'.
u m U1 r ,.
bones, see Tables for brief
.
Warapu, ancestral skulls from, 87 notes on, 130-213

OF THE
mmu Mm
of
<
5 2
a
fa a
o H
H £
Field Museum of Natural History Anthropology, Vol. XXV, Plate XXXIV
TWO FEMALES OF MERAUKE

Front and side views
W
oi .2
© -a
H §
>— i,
< c
S 2
Field Museum of Natural History Anthropology, Vol. XXV, Plate XXXVI
wmBBKam
HEAD-HUNTING TROPHIES OF MERAUKE

Photographed in situ
Field Museum of Natural History Anthropology, Vol. XXV, Plate XXXVII
TWO WARRIORS OF MERAUKE AND TWO NEWLY ACQUIRED HEADS

FROM A VILLAGE WEST OF MERAUKE
° s
W -
J S
< 2
S fa
Field Museum of Natural History Anthropology, Vol. XXV, Plate XXXIX
SKULL NO. 43564

Field Museum of Natural History Anthropology, Vol. XXV, Plate XL
SKULL NO. 43564

Anthropology, Vol. XXV, Plate XLII
Field Museum of Natural History
SKULL NO. 43564

Field Museum of Natural History Anthropology, Vol. XXV, Plate XLIII
, ir*^"/\.
:
*%*
SKULL NO. 43564

Field Museum of Natural History Anthropology, Vol. XXV, Plate XLIV
SKULL NO. 43442

Field Museum of Natural History Anthropology, Vol. XXV, Plate XLV
SKULL NO. 13412

Field Museum of Natural History Anthropology, Vol. XXV, Plate XLVII
4d*
SKULL NO. 43442

Field Museum of Natural History Anthropology, Vol. XXV, Plate XLVIII
SKULL NO. 43442

Field Museum of Natural History Anthropology, Vol. XXV, Plate XLIX
SKULL NO. 43423

Field Museum of Natural History Anthropology, Vol. XXV, Plate L
SKULL NO. 43423

Natural History Anthropology, Vol. XXV, Plate LII
Field Museum of
SKULL NO. 43423

Field Museum of Natural History Anthropology, Vol. XXV, Plate LIII
SKULL NO. 13423

Field Museum of Natural History Anthropology, Vol. XXV, Plate LIV
SKULL NO. 13405

Field Museum of Natural History Anthropology, Vol. XXV. Plate LV
SKULL NO. 411405

Field Museum of Natural History Anthropology, Vol. XXV, Plate LVII
A\ '/
SKULL NO. 43105

Field Museum of Natural History Anthropology, Vol. XXV, Plate LVIII
SKULL NO. 43405

Field Museum of Natural History Anthropology, Vol. XXV, Plate LIX
'
SKULL NO. 43455

Field Museum of Natural History Anthropology, Vol. XXV, Plate LX
SKULL NO. 13455

Field Museum of Natural History Anthropology, Vol. XXV, Plate LXII
SKULL NO. 43455

Field Museum of Natural History Anthropology, Vol. XXV, Plate LXIII
W
{
^^
V
SKULL NO. 43455
Field Museum of Natural History Anthropology, Vol. XXV, Plate LXIV
SKULL NO. 43406

Field Museum of Natural History Anthropology, Vol. XXV, Plate LXV
SKULL NO. -1310G

Field Museum of Natural History Anthropology, Vol. XXV, Plate LXVII
SKULL NO. 43106

Field Museum of Natural History Anthropology, Vol. XXV, Plate LXVIII
SKULL NO. 43406

Field Museum of Natural History Anthropology, Vol. XXV, Plate LXIX
SKULL NO. 43562

Field Museum of Natural History Anthropology, Vol. XXV, Plate LXX
SKULL NO. 13562

Field Museum of Natural History Anthropology, Vol. XXV, Plate LXXII
SKULL NO. 43589

Field Museum of Natural History Anthropology, Vol. XXV, Plate LXXIII
SKULL NO. 43589

THE LIBRARY OF THE
1940
UNIVERSITY Of ILLINOIS
UNIVERSITY OF ILLINOIS-URBANA
3 0112 084204624

Hambly 1940 Craniometry of New Guinea

Hochgeladen von

Dokumentinformationen

Originalbeschreibung:

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Hambly 1940 Craniometry of New Guinea

Hochgeladen von

Copyright:

Verfügbare Formate

W2.

CRANIOMETRY OF NEW GUINEA

THE LIBRARY OF THE

CENTRAL NEW GUINEA

THE LIBRARY OF THE

II. Technique of Measurement 90

III. Age and Sex in the Collection 122

IV. Grouping of the Data 126

VII. Comparative Study of Craniometry of New Guinea and Adjacent

TEXT FIGURES PAGE

Map of New Guinea

Composite Contour Drawings of Skulls of New Guinea (natural size)

Average Contour Drawings of Skulls of New Guinea Compared with Those of

Students interested in the general geography and ethnology of

methods were used for determining the significance of the group

(1) Studying vital statistics by noting the ages of all skulls

Provenance of Skulls from Northern New Guinea

been worked out. The study includes a comparison of small

Provenance of Skulls from Southern New Guinea

and Morant (1933), and if a student is in doubt about the technique

is given for ready reference. To aid comparative study the measure-

List of Measurements and Indices Taken on the Skulls of New Guinea

S = Sagittal arc, nasion-opisthion from opisthion to basion

and the most prominent part of 100 B/L = Cranial index

likely when measurements are retaken by the original

is also desirable that the same series should be remeasured by an-

possibility of using the technique of filling with mustard seed more

accurately than most observers have believed. The use of sand or

Before leaving the subject of technique there is a further point

developing a quick technique for measuring the capacity of skulls.

which was run from a graduated vessel. But, generally speaking,

Calculating Cranial Capacities. — The best known of the formulae

(1) Isserlis' formula, based on measurement of male Negro

(2) Lee's general formula, based on the measurement of male

(3) Hooke's (1926, p. 33) formula, based on a study of male

(4) Bonin's (1934, p. 14) formula for estimating the capacity

(5) Manouvrier's formula as used by van den Broek (1923a,

The measured capacity of 124 male crania from New Guinea

Table of Calculated Cranial Capacities

and 1 cc, respectively, between measured and calculated means,

Hooke's formula gives a rough approximation to the measured

Let us now apply the formula of Isserlis to estimate the capacities

Cranial Capacities by Formula of Isserlis

age measurement of 1.0 mm. or less.

The remeasurement of facial angles may be regarded as satis-

We have, however, to note two significant errors; namely, SC

Possibly an improvement in accuracy in measuring depth of the

be attributed to using a new instrument. In our first test of measure-

i-l <M <M 1-1

co io «o to ffl !C ffl ffl

lO T-i os tj* <* co

P3 05 0)0 0)0 o> a &* a

i-H i-H © 00 <M i-i © -* oo •<* oo ©©

,—,00 ,~.C0 <M 10

^r- 1 ,-~.00 00 <M

o Oi •—-oo ~~'oo _i_

vOO -—sOO O <M

III. AGE AND SEX IN THE COLLECTION

Determination of age was based on data published by T. Wingate

Ages of the Skulls

The population is admittedly small, yet certain definite trends

A noticeable decline in mortality takes place in the age period

Combined Northern and Southern Groups. Inspection of the —