Anthropology. - Cantributians ta the anthrapalagy af the Near-East.

v: Kurds. Circassians and Persians, By C. U. ARIËNS KAPPERS,

(Communicated at the meeting of April 25, 1931).

Of the names written in the title of this paper. only the first two -
the Kurds and Circassians - may be considered as racial names. This
does not of course involve that the groups they indicate at the present time
are pure representatives of such races. or that they are the only represen-
tatives of their race. At best it means that racial characteristics may be
still revealed by those groups.
The last name indicates a geographical and national group. H. and
in how faro it is possible to disentangle its constituents will be discussed
below.
For this a study of the first mentioned races may precede.
At the present time most Kurds live in Kurdistan. a territory belonging
partly to Persia. partly to Turkey. lying approximately between the 35°
and '40° N. lat .. including the Ararat slopes. and extending West to the
Antitaurus in Middle Turkey. and East to Hamadan in Persia (i.e.
approximately between the 37° and 47° E. long). Besides. a large Kurdish
colony. originating from S.E. Turkey (the Kom~agene. Diarbekr and
Mardin ) is settled in Damascus. It is this colony that I examined. with
Mr. MIRZO. a student of the American University of Beirut. and a Kurd
himself. to whom I wish to express my thanks for his great help.
The literature about the anthropology of the Kurds is somewhat contra-
dictory. as appears from the various data concerning their cephalic index.

POLAK 1). describes them as a Nordic race. DUHousSET 2), who measured live Kurds.
found them to be brachycephalic (86.2).
CHANTRE 3). who examined 332 Kurds (272 males and 60 females) from Kurdistan

1) POLAK. Persien. Das Land und seine Bewohner. Brockhaus, Leipzig 1865. p. 18.
2) DUHou~SET. Ëtudes sur les populations de la Perse. Paris. 1863. Quoted from
KHANJKOFF. The origlnal not being at my disposal. I cannot say where the Kurds measured
by DUHoussET lived.
3) CHANTRE. Recherches anthropologiques en Asie occidentale. Arch. du Musée
d'histoire nat. de Lyon. Tome IV, 1895. See also CHANTRE: Aperçu sur les charactères
ethniques des Ansariés et des Kurdes. Bull. Soc. d'Anthrop. de Lyon. Tome I. 1881-1882.
p. 165 and CHANTRE: Rapport sur une mission scientifique dans I'Asie occidentale et
spécialement dans les régions de I'Ararat et du Caucase. Arch. des missions scientifiques
et littéraires, 3ième série. Tome X. 1883. In this paper (p. 40) CHANTRE emphasizes the
variations in the Kurdish index. according to the people amongst whom they live.
I am greatly indebted to Dr. CLAUDE GAILLARD of the Muséum des Sciences Natu-
relles de Lyon for the loan of these valuable papers.
 532

and Caucasia. found an average index of 78.53 (with variations from 70.04 to 86.4)
for the males. A similar male average (78.48) was found by NASSOSSO F 1) with 25 Kurds
from Transcaucasia (Airiga) . IWANOWSICY 2) gave them an ave rage l.w. index of 77.6.
Also VON LUSCHAN 3) stated a prevailing mesocephaly. With lIS adult males from
Karakush he found an index varying from 71.3 to 78.5; with 28 females from Nemrud
Dagh 72.3-78.3. and in Sendshirli the Kurdish index varied from 74.4-80.9. Whereas
all these authors. except DUHOUSSET. agree in giving the Kurds an ave rage mesocephalk
index. PITTARD 4) gives them an index of 86.49. without stating which Kurds he
measured. Such a high index. however. was found by CHANTRE only exceptionally.
with 5 Kurds from Batoum (88.1) and 5 Kurds from Lake Urmiah (86.68) . In Damascus
I measured only 8 Kurds with an index higher than 86.

CHANTRE (1895) is the only author who expressed all his Kurdish
indices in a frequence curve. I reproduce his curve. putting my own
frequency curve of 106 male Kurds. measured at Damascus. underneath
it. As with the Armenians (see my first contribution) our results agree.

70 71 72 '3 74 75 76 7? 78 79 80 81 82 83 8' 8S 8& 87 88 89 90 91 92 93 94 95 '6

Fig. 1. Frequency curve of 332 Kurds (272 r1 aDd 60 2) by CHANTRE.

1) NASSOSSOF. Comptes ren dus Soc. d·Anthr. de Moscou. 1890; quoted from CHANTRF..
(1895. p. 102) . the original not being at my disposal.
2) Quoted by R. MARTIN : Lehrbuch der Anthropologie. Iste Ausgabe. p. 672.
3) VON LUSCHAN. Völker. Rassen. Sprachen. Berlin 1923. p. 91.
4) PITTARD. Race and History. London 1926. p. 366.
 533

In CHANTRE'S Kurdish curve as weIl as in mine the highest top is found

at 78. but in both curves th ere is a large group. following on th is typical
I

or- I ._~-
I
19

18

I?
I
I
I
-
-- _..r- t-- - - H
.~
16
I . 1->'-- r- I- .-
1--' - _. I--
7
14 - -r=- - t-- f- - --

12
i\ i
\
'PI -
I
/ t-- - t-- f-
\Ij
10 - +--

1\
\ ,
I

1\ 1 1
\ I I
1
I
1\ 1
11\ 1- /_\ IN/i\

h~
V
~ht?
I
i
L. ..1_ . _L. __ _ ..i 1 . .
.. '" a;'" '" ..
... ... '"-:.'"
t.'"
a>
'" ~ '"ai g
U>

'"'" M'""-.., '"~ '"

a>
~ '" '";e
.,;
U>
'"
'" :;: ;g ~
U> U>
..; .,;

..
.,;
'""-
..,.,::l ""
R
ai
~
0
a; :;i
i" r- "- ": "" in "" "" "" 0 ..,~
0
"- re OJ
"- "- ~ '""- '""- "
"- """- ...'" ., a; '"""
0
"
'" '" '" "" """"
~ ~
'"
'" '"
ij, '"
'" '"
~
U>

Fig. 2. Frequency curve of 106 male Kurds Erom Damascus.

top. between 81 and 83.9 (or 84 in CHANTRE ' S curve). We further agree
that an index figure higher than 84 is an exception with this people (see
above).
While the most typical top is at 78-78,9 in both curves. the character
of the additional higher index group is more pronounced in my curve
than in CHAN TRE ' S, the additional elevation in my curve lying between
81 and 82.9. As all the individuals registered in my curve we re males. this
extra top cannot be due to sexual difference.
It is evident that both curves - CHANTRE'S as weIl as mine - indicate
the presence of another race mixed with the Kurds. a subbrachycephalic
race with an index varying in CHANTRE ' S curve between 80 and 84. in my
curve between 81 and 82,9 chiefly 1).

1) The 83.87 index was found by CHANTRE specially with the Bilikani Kurds, that
live near Erivan in the Caucasus. CHANTRE is again inclined to consider this as aresuIt
of compression but the northern location of this group is ·s trongly in favor of Georgian
influence.
 534

The index of this additional race suggests that the Kurds are mixed
with another Caucasian group. probably wtth the western Caucasians 1) ;
GIUFFRIDA RUGGERI's 2) Homo indo-europaeus brachimorphus. especially
HADDONS Homo Georgianus 3 ).
This Georgian (or Kartvelian) race is split up in various groups. but
with all of them indices between 81 and 84 prevail. So with 17 Ossetes
CHANTRE 4) found an average index of 83.1; with 12 Georgian Mingrelians
83.2. Similar figures are registered by KHA NIKOFF (83.7) and by VON
ERCKERT 5) . To the same group the Circassians or Tscherkesses belong .
Of thds group 1 measured 54 adult males. Their average measurements
and indices are:

Group I. w. h. I. w. I. w.h.i. I. h. i.

All Circassians 18.58 15.5 13.03 83.42 84 .06 70 . 12

IE. however. I exclude the Circassians whose l.w. index was 87 or

more 6). the average measurements and indices are:

Group I. w. h. I. w.I. w. h. i. I. h.I.

Pure Circasslans 18.83 15.45 13 .07 82.05 84.6 69.41

A similar average 1.w. index (slightly lower) is found with the 30 Cir·
cassians registered by VON ERCKERT.
The curve of all my Circassians. superposed on my Kurdish frequency
curve (fig. 3) shows that the greatest frequency of the index with the
Circassians coincides with the additional group in my Kurdish curve and
makes it probable that the present Kurds may contain a large contingent
of this or of another related subbrachycephalic Caucasian Indo-European
race. thus explaining the controversies in literature concerning the l.w.
index of the Kurds . the original stock of which - as was rightly pointed
out by VON LUSCH.A, N - has to be considered as a mesocephalic race.

1) In the Eastem Caucasus. with the Tschetschenses and Lesghians a higher inde"\:.
occurs. These groups are more mixed with Mongol blood.
2) GIUFFRIDA RUGGERI. Schema di c1assificazione degli Hominidae attuali. Arro. per
I'Antrop. e. I'Ethnol. Vol. 42. 1912. p. 136.
3) HADDON. The races of man. Cambridge Univ. Pre ss. 1924. p. 86.
4) CHANTRE. Mission scientifique dans I'Asie occidentale et spécialement dans les
régions de I'Ararat et du Caucase (I.c. p. 44 and 54).
ö) C.f. also VON ERCKERT. Kopfmessungen Kaukasischer Völker. Archiv. f. Anthro-
pologie Bnd. 18. 1890 (p. 263. and 297). Bnd. 19. 1891. (p. 55. p. 211. and p. 331).
6) 1 found these high indices only with ten men of the Circassian colony at Jeni-Chehr
near Aleppo and suspect Turkish admixture.
 535

The Circassians (or Tscherkesses). however. belong to the subbrachy~

cephalic Caucasian race. Many of them have fair hair and blue eyes. two
features also more frequently found with my subbrachycephalic than with
my mesocephalic Kurds.
~:

2. r---
-
I.
"
I-- I -
17

"

"
'4

"
12

• "l\
•• 1 1\
\/ \
••
.-.\ .. ,
. ..
... ". "• !
... I~
.·· ." . . /\
/ .'
/ " • •• / \

.R' .
•

'"~.
..
..
r
/
. . .;e
\V

:::,.:'" "'. '"~ '"g . ..

.-f-

.
I-- _.

r-
'"
'"
••

. . ." ..g; .. . i.. .. .

D
I-- -

'" ..
. .. '" .:
~. .,; .,; .,;
~ ~- ~
~.
~ ~ ~
N
1;3 .;;
0
~
"
;:: .
" M
"- ~ .
":
~
!!' "-
g: ~
;0
in ~
M
ro ;f,
<0
on
<»
,.: '"
'" '"'" '" ""'" '"'"
.
0 '" .;; ~ .. ..
M ~
:!l .
Fig. 3. Superposltion of the frequency curves of 106 cf' Kurds (crossed curve).
and of 54 cf' Circassians (continuous curve).
Approaching with these data the anthropology oE Persia. I should start
by emphasizing that two thirds of the 165 Persians enlisted in my dossier
came from that region of Persia that lies North of the line Hamadan~
ISphahan. One third only came from more Southern parts of Persia. from
S .W. as weIl as Erom S.E. Persia.
In Chusistan. S .E . from Bassorah (in Mesopotamia) and in Laristan. East of the Persial1
gulf. the population is said to contain many Arabic elements. Hence the name Arabistan.
sometimes given to this region.
This is not strange since this part of Persia continues gradually in Mesopotamia
without aseparation by such high mountainous ranges as occur more North. Also the
Persian gulf proYides an easy means of intercourse between Laristan and Arahia
proper 1) . Among this Semitic population there may be Adnan Arabs as weil as Kohtan
Arabs.

1) It is an interesting fact that water that giyes the sharpest geographical distinctions.
so frequently causes an anthropological expansion and intermixture. This factor apparently
 536

I leave tbe Semitie eonstituents (also the Persian Jews, studied by WEI SS ENBE RG ) , and
also the Mongol elements out of eonsideration here, being the least important numerieally.

Of the 165 Persians enlisted in my dossier, 9 males and 41 females we re

measured in Persia by Miss ELEONOR BUSH for Prof. H ARALD KRISCHNER
of the American University at Beirut, who kindly put these figures at my
disposal. I am much indebted to Miss BUSH and Prof. KR!S CHNER for
their help.
The remaining Persians I measured at Beirut with the help of Persian
members of the staff and Persian students, always controlling the Persian
origin of my material.

19 t--t---t--If---t---t--I---t--+-j-+- + ----t--f- l l--t---t-----,I--t---+-jf--+---f--- _+_-+-1_+---+--1

1
1
\ .
I •
I1 •• •
•
1
•
1
1 •
~
1 • ·
••
-- ..• •
.
'"•
\
••
••

~ .
;g !;i'"
.. '"R . . oi ,.,.;'" ;1
.. "
.,;
.
" '" l' g'" .. .. " .. " .
;;;'" ;g '"~ ~'" .. '" '"
. '",.;
"~ re~ ~" ~ R ~
~ $ ~ N
f, ~
'"
N'
~ .;;
" '" "? ij;
~
'"
:;: '" " ~
N M
0
:;;
" "'"
N

"
0
'" ~ .;; i'2 J, '"'" '" '"
'"'"
0 N M

'" '" '" " " t"-

'" '" 00 ii.
'" '"
Fig. 4. Continuous line: curve of 165 Persians; erossed line: curve of 106 Kurds.

was of still more importanee in ancient times, wh en traffie by water was more easy than
traffie by land. The same is observed with the spread of the Phoenicians and Greek
round the Mediterranean.
Mountainous ranges, on tbe other hand, frequently are sharp anthropologieal boundaries.
the hest example being the Karakorum and Hindoo-Koosh, North of whieh hyperbraehy-
cephalie Mongols and South of whieh the mesoeephalie Indo-afghans live. (cf. also
UJFALVY, l.c. infra p. 52, and RIPLEY l.c. infra p. 452 : "from either side purity of types
of very different sort inereases to the watershed whieh lies between them.")
 537

I am especially indebted to Mr. ZAYNE NUR~ED~DIN and Mr.

DOWLATSHAHI for their help.
Though all the people enlisted in my dossier called themselves Persians.
the superposition of my Persian and Kurdish curves (fig. 4) shows that
the typical Kurdish top (78-78,9) returns with the Persians in a striking
way.
Also the subbrachycephalic Caucasian race is present again with the
Persians, as appears' from the elevations at 81-81.9 and 83-83,9, th is
index occurring also with the Georgians and Tscherkesses.
The 86-86,9 top of the Persian curve may be caused by Armenian and
Armenoid influence 1), especially by the Armenian group of 86,53
(-+- 1,49; cf. my second contribution ).
The Northern origin of the subbrachycephalic Caucasian and hyperbrachycephalic
Armenoids may explain the fact that these indices were special1y evident with my
Pcrsians, originating from places North of the line Hamadan - Isphahan, whereas South
of that line only 9 % of the individuals measured had an index higher than 81 2 ) .
Now considering the Ie ft hand si de of my Persian frequency curve we
see that there are a great many subdolichocephalics in it as is shown by
the high elevation at 74-74,9 and some lower on es at 67-68,9 and 70.
These e1evations, chiefly caused by Southern*Persians, indicate another
component of the population: the Indo*Afghan component, while in the
North a Turkoman element may contribute to the 74-74,9 elevation.
Personally I only examined six adult men, who called themselves
Turkomans, tall Indo-European looking people. Their measurements and
indices are so charaç::teristic that I cannot omit giving them here:

Turkomans length wldth helght 1. w. I. I w. h.I. I. h.I.

No. 1 20.0 li.5 13.2 72.5 91.0 66.0

1
No. 2 19.1 li.l 12.9 73.8 91.5 67.5
No. 3 19.6 14.0 13.2 71.4 91.3 67.3
No. 1 19.0 11.1 13.2 76 . 3 91.7 69.5
No. 5 19.0 11.0 12.0 73.7 85.7 63.1
No. 6 18.6 11.7 12.2 79.0 83.0 65.6

Average. 19.221 11. 28 1 12. 78 1 71. 15 1 89. 53 1 67.3

1) As already mentioned in my first contribution Abbas the Great (1605) established

an Armenian colony at Isphahan, some members of which have kept their Annenlan
characterlstics though many have obtained a mesocephalic index, depending probably
on marriages with other races.
2) These came from Kermandshah and Sjiraz, i.e. from the South-West. Those above
&4 may due to Mesopotamian influence (Kohtan Arab influence), wihile some higher
illdices (87-89) may be due to Bakihtyari influence (see below).
 538

The data concerning the Turkomans are contradictory. DUHOUSSET.

who measured one male Turkoman from Khiwa. found an index of 79.1.
JAVORSKY 1). who measured 59 individuals. found very similar figures:
an average length of the head of 19.3. an average width of 14.6. thus
making an average l.w.i. of 75.6 (ranging from 68.69 to 81.76).
DENIKER 2) gives the Transcaspian Turkomans an average of 77.9
(23 individuals ).
Whereas these authors agree that the Transcaspian Turkomans are
mesocephalic. RIPLEY 3). following UJFALVY 4) gives them a very high
brachycephalic index (85-89) and considers them to be an alpine race.
Apparently the divergence of opinion between DUHOUSSET. JAVORSKY.
DENIKER. and myself. on one hand. and UJFALVY. on the other. depends
on what race should be considered as the actual Turkoman race.
Turkmenia proper is only a small part of Russian Turkestan. viz. that
part that lies between thc Caspián and Aral seas. bordering on Persia.
IE. however. the name Turkomans be applied to the population of who\c
Russian Turkistan. extending as far east as Chinese Turkestan. as some
anthropologists do. it would include also the Uzbegs. Kiptchaks and even
the Kirghese that are reported to be highly brachycephalic (ind. 84-85).
the Kirghese being. moreover. evidently Mongols.
T.he different character of many so~called Turkomans has already been
stated by KHANIKOFF 5) and DUHOUSSET (I.c. ). the former giving the
Turkomans of Persia an average index of 82.
IE DENIKER and JAVORSKY are right. as the figures I found with my six
Turkomans suggest. then the Turkomans proper are meso~ or dolichoce~
phalic and then the Northern elements in the dolichocephalic top of my
Persian curve may include people of Turkoman origin. belonging perhaps.
to the Khodjar (or Qojar) and Afshar Turkomans living in North~East
Persia and from which also some of the former rulers of Persia deseended.
Leaving aside the rather contradictory data concerning the Turkomans.
there is no doubt that the majority of the high 73-74,9 top in my
Persian curve is caused by Indo~Afghans . to which also a part of the popu ~
lation of Loristan may be added. For the Afghans DE QUATREFAGES and
HAMY 6) found an index of 73,15 a figure also stated by HOUSSAY 7) with
some Loris (73.57). As the main index of the Hindoos proper, according to

1) ]AVORSKY. Mil. Acad. Anthrop. Soc. oE Petrograd 1895 Vol. 11. p. 145. quoted
Erom RIPLEY and DUDLEY BuxTo N, 1897. See also ]AVORSKY p. 193, where thls Index
oE 75.6 Is given as an average oE 191 measurements.
2) DENIKER. Les peuples et races de la terre. p . 669. Paris, 1900.
3) RIPLEY. The peoples of Europe, London, 199 p. 416 a.E.
4) UJVALVY. Les Aryens au Nord et au Sud de l'lndou-Kouch, 1896 (Quoted Erom
RIPLEY. the original not being at my disposal) .
5) KHANIKOFF. Mémoire sur J'Ethnographie de la Perse. Martinet. Paris. 1866.
6) Cranla ethnica. Quoted Erom HOUSSAY, the original not being at my disposal.
T) HOUSSAY. Les peuples actuels Ie la Perse. Bull. Soc. d·Anthrop. de Lyon. Tome 6.
1887. p. 101.
 539

HOUSSAY is between 72.28 and 74.78. according to DUHoussET 74.5. the

conclusion that these groups belong together is likely to be true.
Whether the Hadjemis (index 77.77) and Tadjiks may be a branch of this group. as
is asserted by CHANTRE (1895. p. 174). I dare not say. but there may be a close relation
between the dolichocephalic Turkomans and the Djemchidis Afghans; DENIKER l.c.
p. 438). the more 50 as the Turkomans I measured also had a distinct Indo-European type.

Another group of the same race may be the Zoroastrian Persians Erom
Farsistan, formerly very numerous in S. Persia, where they were described by
KHANIKOFF. (I.c.).
Personally I only have the measurements and indices of two Persians
from Fars. They are 14.7 by 19.2. i.e. 76.6 for a male. and 14.4 by 18.3,
i.e. 78.8 for a female 1).
50 much concerning the racial analysis of the Persians which is not at
all complete. but still may add a Iittle to our knowledge of the anthropo-
logy of th is country where hitherto so few researches have been made 2).

I cannot omit entering here upon another question. viz. the question of
thc origin of the Druses. sin ce recently HITTI in a very interesting paper 3 J
advanced a number of historical and Iinguistic arguments opening the
possibility th at the Druses might be related to the Persians or to the
Kurds. As far as concerns the Kurds, anthropology. however. gives no
arguments for this supposition. as the ultrabrachycephalic relations of thc
Druses are very rarely met with amongst the Kurds and certainly are not
typical of th is race (see above).
Besides the word "Persians" is a geographical expression, unless one
would consider the Farsis as the real Persians. descending perhaps from
the ancient Perses.
The index with the Persian Farsis, however. being dolicho- or mesoce-
phalic. and the one with the Bombay Farsis being subbrachycephalic. I
cannot see an anthropological relation here.
Taking the expression Persians in the widest or geographical sensr
there is only one sm all top in my Persian curve with a hyperbrachyce-
phalic index, but the index of this top (86-86.9) is so typically Armenoid
that there is a great probability th at this elevation is caused by Armenian
influence. easily explained by the immigration of this people into Persia,
already favored by Abbas the Great.

1) It is weil known that most Farsis left Persia after the decline of the Sassanids.
migrating to India. The Farsi in Bombay. however. are not at all dolichocephalic. but hav<:
an index of about 82. according to DENIKER (l.c. p. 672). Also VON LUSCliAN (l.c.
p. 111) stated that the index of the Indian Farsis is different.
2) The data given in OE MORGANs great work on Persia have more ethnological than
anthropological value (Mission scientifique en Persie; 1894).
3) PH. K. HITTI. The origin of the Druse people and religion with extracts of their
sacred writings. Columbia University Oriental studies. Vol. 28. 1928. I am greatly in-
debted to Mr. SULAYMAN ABu IZZ-ED-DlN of Beirut for calling my intention to Dr. HITTl'S
paper.
 540

Curiously this top coincides (see my superposed Persian and Druses

curves in fig. 5) with a group amongst the Druses. which I also consider as
Armenoid or -if the Armenoid type is the Hittite type - as Hittite.
About half of the Druses. however. have a higher index than 86.9.
Wh ether this be due to the inbreeding amongst the Druses. combined with

cr>
~ .'" ~. ..
cr>
.,;
~ ... .~
r<

'"cr> ~
cr> .
cr>
,;,
'"cr> r'- cr>

Fig. 5. Frequeney curves of 116 r! Persians (broken line) and of 84 r! Druses

(continuous hne).

the hereditary dominanee of brachycephaly. or with external influences

exercised on the skull in the first years of childhood. I leave out of con-
sideration here. So much is sure that the other elevations in my Druse
curve represent indices (88-91.9) occurring only with 5 % of my
Persian curve (fig. 5). These ultrabrachycephalic Persians came from
various places. but those from Kermanshah may have been Bakhtyari.
with which DUHoussET, who measured four of them, found an average
index of 89 (KHANIKOFF l.c. p . 138). Also GAUTHIER (quoted by CHANTRE,
1895, p. 136), who measured nine Bakhtyari, found an average Lw. index
of 89.321).

1) DENIKER, l.c. p. 486 gives t,hem an average index of 84.5 and seems to group them
togE<ther with the Loris as does KEANE (l.c. p. 279). There are, however, meso- and
brachycephalic Lons, and the latter may be Loristan Bakthyari.
HOUSSAY, who measured only three Bakhtyari, found an average index of 83.7.
 541

The Bakthyari are a short (Iength 1.67 M.) mountainous population in West-Persi'l
(Loristan). South of Kurdistan. They mostly speak Kurdish. but are not to be
considered as Kurds. Excellent pictures of this people are given by DUNLOP.1)

The fact that several of the present Lebanese Druses may have had an-
ces tors in Iraq on the border of Persia is no argument for their Persian
des cent. as especially Iraq even now contains many Kohtan Arabs. among
them highly hyperbrachycephalic individuals. And where VON LUSCHAN
stat es In his Huxley memorial lecture on the early inhabitants of Western
Asia that of fifty nine skulIs of adult male Druses measured by him. "not a
single man feIl - as regards his cephalic index - within the range of thc
real Arab" ~). he apparently considers here only the Adnan Arabs. not
the highly brachycephalic Kohtan Arabs. so numerous in Iraq.
Unhappily the bloodindex of these Kohtan Arabs is unknown hitherto.
IE. however. it is the same as with the "Moslim Arabs" examined by PARR
it would be 1.59 i.e. the same index as found by P .'RR for the Druses.
while. with his Persians. PARR found an index of 1.35.
Whatever may be the truth about the Kohtan bloodindex. anthropology
hitherto does not afford any argument in favor of a specific Persian
relation of the Druses 3) unless one would be inclined to consider the
Bakthyari as their relatives. for which. however, more researches would be
necessary.

1) H. DUNLOP. Perzië voorheen en thans. Firma Erven Bohn. Haarlem. 1912. p.

566--568.
2) Which of the ma in Arab groups is the "real one··. the Northern mesocephalic
phoenico - palmyrene Adnan Arabs. or the Southem brachycephalic (Iraqquian and
Yemenite) Arabs. is a matter of appreciation. Contrary to VON . LUSCHAN. the Arab
tradition considers the Yemenite group (el ariba) as the more "real"' one and not the
Adnan group (el must"ariba. i.e those who ca me later) C.f. HOUTSMA. ARNOLD. BOSSET
and HARTMANN Encyc10paedia of the Islam. Part I; Brill. Leiden; Harrassowitz, Leipzig,
1913.
3) Linguistic arguments are of Iittle value in anthropology. Many examples may be
quoted of racially different peoples using even the same speech. Besides Persian ex-
pressions occurring with a people whose ancestors Iived in Iraq about l000-boo A.D.
would not be strange considering the Persian influence on Iraq before that time (during
the Sassinide empire) and the union of Persian and Mesopotamian territory under the
caliphate of Baghdad. under which the Druse sect originated.

35
Proceedings Royal Acad. Amsterdam. Vol. XXXIV, 1931.