515
Documenta Praehistorica XLIII (2016)
Archery by the Apaches – implications of using
the bow and arrow in hunter-gatherer communities
?iga {mit 1,2
1 Faculty of Mathematics and Physics, University of Ljubljana, SI
2 Jo/ef Stefan Institute, Ljubljana, SI
ziga.smit@fmf.uni-lj.si
Introduction
The invention of the bow and arrow provided the
early hunter with a remarkable hunting tool. How-
ever, in complex human relations, it also turned into
an efficient weapon to be used against others. In
the prehistoric Old World, various testimonies to
human violence exist, most of them manifested in
changes in human skeletons, although some expli-
citly point to injury caused by an arrow. The earliest
evidence of armed conflict from the Late Paleolithic
(12000–10000 BC) is from Jebel Sahaba in Sudan,
where half of 107 burials suggest violent death (Chri-
stensen 2004.134). Projectile injuries in the Middle
and Upper Paleolithic are rare, but in the Mesolithic
they amount to 17 out of total of 77 registered exam-
ples, or 22% (Estabrook 2014.65). A review by the
same author (Estabrook 2014.53–58) lists the fol-
lowing Mesolithic sites with projectile bone injuries:
Bøgebakken in Denmark (two perimortem projec-
tiles), Vlasac (one perimortem bone projectile) and
Schela Cladovei in the Danube region (6 perimor-
tem flint or bone projectiles), and Vasilyevka in Ukra-
ine (8 perimortem projectiles). Regarding the Vlasac
example, the author also admits the possibility that
the wound was inflicted by a dart (Roksandi≤ 2006b.
342). Further examples of Mesolithic death caused
by projectile injuries (Christensen 2004.134–135)
include Téviec in France (two arrowheads in the
spinal column) and Henricksholm/ Bøgebakken in
ABSTRACT – This review focuses on the technical and social details of production, training, and use
of archery equipment by a Native American tribe, the Apaches. The study aims to understand the use
of the bow in the Mesolithic and Early and Middle Neolithic societies of the Old World. The paper fur-
ther describes arrow ballistics. An arrow and bow with similar dimensions and materials to those used
by the Apaches was reconstructed and used in ballistic experiments. Shooting and the subsequent mo-
del calculation showed that the effective range of arrows made of reed and projected by a bow of
medium strength (16–18kg) was not more than approx. 20m. Due to the initial flat part of the ballis-
tic trajectory, such arrows were quite efficient in close-range contests. Within the model calculation,
a regression procedure was introduced to determine the arrow air-drag parameters from an ensemble
of shots.
IZVLE∞EK – ∞lanek podaja pregled o tehni≠nih in sociolo∏kih podrobnostih pri izdelavi, treningu in
rabi lokostrelske opreme pri ameri∏kem domorodnem plemenu Apa≠ev. πtudija je namenjena razu-
mevanju rabe loka in pu∏≠ic v mezolitskih ter zgodnje in srednje neolitskih kulturah Starega sveta.
Drugi del ∏tudije se ukvarja z balistiko pu∏≠ice. Predmet balisti≠nih preizkusov sta bila pu∏≠ica in lok
podobnih dimenzij in materialov, kot so jih uporabljali Apa≠i. Streljanje in modelsko ra≠unanje sta
pokazala, da uporabni doseg pu∏≠ice, izdelane iz trstike in izstreljene z loka srednje mo≠i (16–18 kg),
ni ve≠ji od pribli∫no 20 metrov. Ker je balisti≠na krivulja take pu∏≠ice v prvem delu leta zelo polo∫-
na, so bile pu∏≠ice izredno u≠inkovite na kratko razdaljo. V modelskem ra≠unu smo uvedli nov po-
stopek za dolo≠anje parametrov zra≠nega upora iz skupine strelov.
KEY WORDS – Apache; bow and arrow; reed arrow
DOI> 10.4312\dp.43.28 
?iga {mit
516
Denmark (bone point in the throat). From the Meso-
lithic and Neolithic, we have further evidence in gra-
phic depictions showing armed contests between
archers. In Les Dogue, Ares del Maestra, eleven ar-
chers are attacked by seventeen running archers,
and in Cueva del Roure, Morella, four archers are in
the pincers of three others (Christensen 2004.135).
In a mass grave at Talheim in Heilbronn, dated to
the Early Neolithic, two individuals among 34 buri-
als show injuries caused by arrow points. From the
Middle and Late Neolithic, finds of human bones
with embedded flint or bone tipped arrows are nu-
merous: Castellet, Villevénard and Roaix in France,
Porsmosen in Denmark, the Stapleton enclosure
near Dorset and Stonehenge in the UK (Christensen
2004.136–137), Aktopraktic in Anatolia (Erdal 2012.
80). Two projectile injuries out of a total of ten trau-
matic injuries are also documented for Paleoameri-
cans in North America (Chatters 2014.75). 
The intensity of conflict may be estimated from the
percentage of individuals showing signs of violent
death in a certain grave assemblage. In Schela Cla-
dovei, in the area of the Iron Gates, 15% of indivi-
duals died violently from shots of arrows tipped
with bone points (Bonsall 2008.263). On the other
hand, Mirjana Roksandi≤ (2006a) estimates that con-
flicts were occasional, as among the 418 individuals
she studied, only a few showed the consequences of
violence. An opposite view is given by Jonas Chris-
tensen (2004.130): armed conflict in primitive socie-
ties caused 10–40% loss of population. For Linear
Pottery culture, the skeletal material is well docu-
mented and shows that 20% of the population died
violently (Christensen 2004.152).
As material remnants provide only limited sources
of complex human relations, behaviour patterns are
also studied by ethnographic research on contempo-
rary, but archaic, societies which are still attached to
their traditional way of life and have survived in a
few remote parts of the world. The cultural groups
most studied today or in recent decades are hunter-
gatherer communities in Alaska or Australasia (Kra-
mer 1979.3; LeBlanc 2014). However, with modern
communication technologies, their life-style is rapid-
ly disappearing, and many interesting data can be
found only in the literature published since the sec-
ond half of the 19th century. The present work aims
to study the use of the bow by the Apaches, an in-
digenous tribe in the south-west of North America,
about whom significant data exist in their memor-
ial literature from the end of the 19th century and
in ethnographic studies published between 1930
and 1960. In a review, we would like to give a com-
prehensive view of the technology of making archery
equipment and the use of bows in hunting and com-
bat. In the second part, we try to reproduce some
specific item of archery equipment, notably reed ar-
rows, and test their ballistic properties.
The Apaches, the Athabaskan-speaking native pop-
ulation, settled in the area of present-day Arizona,
New Mexico and the northern borders of Chihuahua
and Sonora between the early 15th and 16th cen-
turies. Among this large group, we limit our study
to the Chiricahuas and Western Apaches. The main
memorial literature on the Chiricahua is the account
of Geronimo as recorded by Stephen M. Barrett in
1906 (1996), and the autobiography of Jason Betsi-
nez (1959). Many observations can also be found in
the works of Eve Ball (1988; 2003). For the Western
Apaches, ethnographic material and memories of
war parties from the second half of the 19th centu-
ry were recorded by Grenville Godwin and are avail-
able in Keith H. Basso’s book (Basso 1971).
By the first half of the 19th century, the Apaches
were living quite an independent way of life, dur-
ing the period that started with Mexican indepen-
dence in 1810–1821 and ended with the Mexican-
American war in 1846. This period is characterised
by an absence of organised Mexican military control,
while the influence of white Americans was almost
negligible. Relations with Mexicans were complex,
involving peaceful co-existence and fierce military
conflicts (Utley 2012.16–28). The Chiricahua econ-
omy was based on four main activities: hunting, ga-
thering, modest farming, and casual looting of their
Mexican neighbours. During looting expeditions or
raids against the Mexicans, they usually used cold
weapons, as described in the memories of Gero-
nimo. The Western Apache economy was more ad-
vanced, being primarily agricultural; they also built
dams and dug irrigation ditches (Basso 1971.79).
However, it is interesting that they quit the irrigat-
ed cornfields as soon as the corn ears developed,
and temporarily left to collect acorns (Basso 1971.
95–96). The Western Apaches generally made the
transition to reservation life much more easily than
the Chiricahuas. In the earlier period, the Chirica-
huas also organised two kinds of military expedi-
tion: smaller parties of up to fifteen warriors gene-
rally aimed at looting, while a large number of war-
riors led by respected chieftains and a shaman (Bas-
so 1971.16–18) normally avenged the deaths of ca-
sualties.
Archery by the Apaches – implications of using the bow and arrow in hunter-gatherer communities
517
During this period, the bow was used as an effec-
tive long-range weapon, supported by the fact that
contemporary guns were muzzle-loaders superior
to their predecessors of previous centuries in having
a more efficient firing system of percussion caps
only. Mexican military doctrine had remained essen-
tially unchanged since the Napoleonic wars and was
based on the organised concentrated fire of multiple
weapons, rather than on precise individual mus-
ketry. In such circumstances, the bow could compete
with the gun in range and precision, and bettered
it in the rate of fire.
Use of the bow
Hunting
Hunting was the main activity of men and boys as
soon as they passed the age of eight or ten (Barrett
1996.65). The bow and arrow was not the only hunt-
ing weapon; larger animals were also hunted with
the spear and smaller ones with a short stick or a
hunting club. Horses were also used in hunting, no-
tably for catching up with smaller animals such as
turkeys and rabbits. The bow and arrow (but also
spear) from horseback were used to hunt buffalo
(Barrett 1996.64), though for Apaches, the buffalo
was probably an occasional pray. Geronimo careful-
ly describes efforts during deer hunting, which was
based on concealed approach to a grazing herd. A
skillful hunter could get several animals before the
rest of the herd noticed him and ran away. Although
not described explicitly, this type of hunting was pos-
sible only with bow and arrow. A good shot could
kill a deer instantly at a distance of a hundred yards
(90m).
Another animal that Geronimo explicitly says was
hunted with the bow is mountain lion, although he
also says he killed one with a spear (Barrett 1996.
68). The bow and arrow can also be used for some
unusual practices, such as shooting down parts of
a beehive of wild bees (Ball 2003.122).
The bow was used for hunting during the Turkey
Creek settlement period (Davies 1929.102–120) as
late as 1884, possibly due to the scarcity of ammu-
nition (Ball 2003.159). Jason Betsinez remembers
that the men used the bow to kill domestic animals
allotted to them for food (Betsinez 1959.6). He was
also impressed by the penetrative ability of the ar-
row, as he writes, “Many men can drive an arrow
clear through a large animal.” At a distance of
14m, an arrow could be sent through a deer (Opler
1996.388).
Combat
The Chiricahuas used the bow as the main ballistic
weapon until about 1860. It was at the battle of
Apache Pass in 1862 that they first used a consid-
erable number of firearms – although most of them
still carried bows and arrows (Ball 1988.20) – but
they probably lacked sufficient ammunition (Swe-
eney 1991.437). During the 1870s, bows were used
in small-scale conflicts on the Tularosa Reservation
(Thrapp 1974.156). In an ambush set for Mexican
soldiers in 1875 near Fronteras, the Mexican com-
mander was reportedly killed by an arrow (Sweeney
2010.37). Bows were often used in internal conflicts
and scuffles, which regularly followed excess con-
sumption of alcohol (Sweeney 2010.26). Geronimo
describes treating wounds after such fighting (Bar-
rett 1996.96). Betsinez remarks that bows were sel-
dom used after 1882 (Betsinez 1959.85). In spite of
this, Bourke noticed in 1883 during the Mexican
campaign of General Crook that the Apache still car-
ried bows and arrows as a side arm, as well as the
latest repeating rifles and revolvers (Bourke 1958.
102). The Western Apaches conducted raiding par-
ties in Mexico armed with bows until the 1850s or
1860s (Basso 1971.84, 195).
Most Apache raiding parties started on foot, and
each warrior carried equipment and rations for
three days (Barrett 1996.84). From a description of
fighting scenes in Geronimo, we learn that personal
arms consisted of a spear, bow, arrows, and a knife
(Barrett 1996.93), and eventually also a war club
(Opler 1996.340). The number of arrows carried in
important encounters was between thirty and forty
(Opler 1996.388). In three places in his memoirs,
Geronimo says that he ran short of arrows during a
fight. The first two were in fights with Mexicans. In
a supposed revenge battle (probably the battle at
Pozo Hediondo in 1851), Geronimo used all his ar-
rows in about two hours of fighting (Barrett 1996.
62). In some other attempt to raid an unnamed vil-
lage, an Apache party was ambushed and pursued
(Barrett 1996.85). The soldiers subsequently chased
the Apaches, who were fleeing and hiding. From his
hiding place, Geronimo used his bow to defend him-
self against soldiers who had spotted him. According
to his account, he had enough time to take deliber-
ate aim from a concealed place until he had spent
all his arrows. These circumstances do not allow us
to determine the range at which Geronimo used his
bow, but the distances were supposedly from some
tens of metres up to a hundred metres. When he left
the hiding place, mounted Mexicans threatened to
cut him off. In this case, the bow would have been
?iga {mit
518
used at very close range. The third description refers
to a conflict with United States soldiers in 1863,
1864 (Barrett 1996.120). The Apaches were only
modestly armed, so Geronimo soon spent all his ar-
rows. In a fight ten days later, he had spent all his
arrows before ten o’clock in the morning, although
the fighting lasted the whole day.
The bow’s short range made it ineffective against
firearms. On a raiding party near Arispe around
1867, the Apaches were pursued by Mexican horse-
men (Barrett 1996.99). When they opened fire, they
were too far away for the Apaches to respond with
their bows. However, bows had sufficient range to
shoot after the fleeing enemy (Barrett 1996.93). The
bow was effective in short-range combat. This was
seen when Mexican troops attacked Geronimo’s set-
tlement (Barrett 1996.89). In the thick of the fight-
ing, Geronimo was able to shoot an officer and es-
cape. The rate of fire of the bow was an advantage
even in contests with modern repeating guns. As
claimed by an unnamed warrior, it was possible to
shoot three arrows before the opponent could load
the magazine of the gun (Opler 1996.388). Long-
range shooting is recorded for the Mescalero Apa-
ches. In 1868, they besieged Mexican soldiers who
had taken their horses into an enclosure. The Apa-
ches shot their arrows high in the air so they fell in-
to the enclosure and killed all the animals (Sonnich-
sen 1958.144).
Production of archery equipment and training
Archery equipment was manufactured by experi-
enced old men who passed on their knowledge to
the young (Basso 1971.226). Young boys had to
observe men making bows (Opler, 1996.28). If the
older man was a relative, he made the bow for free,
but should then receive a portion of prey or booty
without asking (Basso 1971.226). If the bow maker
and owner were not related, the payment would be
a shirt or buckskin (Opler 1996.226). Men were sup-
posed to make arrows themselves. Making a full qui-
ver of arrows took a month (Opler 1996.388).
Apaches trained to fire the bow from early child-
hood (Oper 1996.50–51). In a Massai legend (Bet-
sinez 1959.143; Ball 1988.248–261), we learn that
a child had to shoot through metal ring approx. 90m
away (Ball 1988.248). In 1869, Mexican troops
caught a man in the Guadalupe Canyon who had
taught a boy how to make arrows (Sweeney 1991.
267). Children practiced bow shooting until the
1880s, when the bow was no longer used as a wea-
pon (Ball 2003.113). Young Chiricahuas trained by
shooting fish in streams, although animals like frogs
and snakes were never eaten (Barrett 1996.68).
Mastering the bow was also developed through nu-
merous archery games; the contests involved shoot-
ing into the distance or touching the arrow shot by
the first contestant. If the arrows touched, the sec-
ond archer could keep both arrows. If he missed, his
arrow served as a new target. A roll of twisted grass
was thrown into the air and had to be hit before it
touched ground (Opler 1996.50–52).
Even in adulthood, knowledge of the bow was main-
tained by frequent competitions. These events also
involved fast shooting. The aim was to shoot three
arrows before the first touched ground (Opler 1996.
391). A ball made of cedar bark was used for target
shooting (Basso 1971.226).
The string was held by the middle three fingers of
the right hand, with the first finger above the ar-
row nock. The bow was held with the left hand, and
the arrow rested on left side of the bow on the ar-
cher’s thumb (Basso 1971.225). The three fingers
could also be used so the string was held between
the thumb and first finger, with the second finger
assisting (Opler 1996.388). For close shots, the bow
was held vertically, but for long shots the bow was
held crossways (Basso 1971.226). Aiming was taken
along the arrow. A wrist guard made of rawhide
(Opler 1996.391) or leather (Basso 1971.224) was
often worn on the left wrist to protect the wrist
when it hit the string (Basso 1971.225).
Making the bow
Apache bows were made from mulberry wood. The
Chiricahuas cut second-growth trees found at foot of
the hills (Betsinez 1959.5). After being cut, the wood
was dried for a long time and then worked to the
proper size, which was a man’s wrist in the middle
and tapering at the ends (Betsinez 1959.6). The
Western Apaches also used mulberry wood, but fa-
shioned the bow when the wood was still green, and
then they hung it to dry for about five to seven
days. The shape of the bow was a simple arc, with
a straight middle section and flexible limbs. The
Western Apaches also used bows in the shape of
double arc, but learned that single arc bows are bet-
ter. The double arch could be formed only when the
wood was still fresh (Basso 1971.223–224). Oak,
locust and New Mexico maple also furnished accept-
able wood (Opler 1996.386).
Archery by the Apaches – implications of using the bow and arrow in hunter-gatherer communities
519
The shape of single-arc bows can be seen from a few
photographs showing warriors with full equipment
(Fig. 1). Comparing the bow size with the dimen-
sions of the human body, the length of the bow with
stretched string was between 1–1.2m. Opler states
“three to four feet” or the “length of two arrows”,
adding “the short or hardwood shafted arrows” in
the note (Opler 1996.386). The bow could also be
smaller for hunting small animals and for training
boys (Opler 1996.388). A size of 1.3–1.4m can be
deduced from the photograph taken in the studio
(Fig. 1); however, the bow here may also be an im-
provised item. Nevertheless, we learn from these
photos that Apache warriors transported and shot
the bow with several arrows in the hand that held
the bow, being thus ready for a rapid exchange of
shots. There are also several photographs of Geroni-
mo making bows (Fig. 2). Some reveal the bow pro-
file, with a width of about 4cm at the grip and a
length of 1m. Supposedly, these dimensions were
traditional, although one should not overlook that
the bows in Geronimo’s photographs were made for
sale to tourists, and that Geronimo was wise enough
to adjust his production to demand. It is known, for
instance, that the Western Apaches made bows for
tourists of lower quality wood (Basso 1971.223).
In the crafting procedure, the branch is cut, split,
shaped and smoothed, and then left to season (Op-
ler 1996.386). The wood was then greased and
made slick and shiny. To bend into its final form, it
was heated in hot ashes, shaped and then tied and
left to dry for about ten days. To shape a double-arc
bow, bending was easily achieved between two
young trees growing close together (Opler 1996.
386).
The outer surface of the bow was painted in a solid
color, while the inner surface was decorated with
stars, crosses, serrated or parallel lines, or naturalis-
tic figures (Opler 1996.386). The inner part of the
bow is seen in a few photographs of Geronimo. In
one of them, we can recognise a zigzag line and on
another a series of triangles (Fig. 2).
The bowstring of the Western Apaches was made of
sinew from the back of the deer or from the mus-
Fig. 1. Photographs of Apache warriors with bows and arrows: two youths at San Carlos Agency in 1888
(Utley 1977), a warrior around 1860 (artificial scenery in the studio), Chatto, chief and later army scout,
with reed arrows and boy with a child’s bow (Fort Sill after 1894).
Fig. 2 Selected photographs of Geronimo posing
with a self-made bow. Top row, left: in Mont Ver-
non, between 1888–1894; Bottom row, left: in St.
Louis, 1904.
?iga {mit
520
cle of the hind legs. The sinew was soaked and split
into strips that were spliced together. The long
string was then doubled and twisted together. The
string was gradually made tighter after it was first
put on the bow (Basso 1971.224). Twisted sinew
was also used for bowstrings made by the Chirica-
huas (Betsinez 1959.6). The warrior usually carried
an extra string in case one should snap. If necessary,
a good string could also be woven from the fibres
of the mescal leaf. When not in use, bows were car-
ried unstrung (Opler 1996.387–388).
Although the Apaches never produced composite
bows with a sinew layer on the outer part of the
limbs, they could use sinew or horse hair to bind a
bow that was weakened by use. The sinew was fixed
with glue made from boiled hooves and horns or
occasionally piñon pitch (Opler 1996. 387). Some-
times pitch was rubbed into the inner side of the
bow to strengthen it (Basso 1971.230). For compa-
rison, the existence of composite bows in the Euro-
pean Neolithic is inferred from a rock carving in a
Corded Ware tomb in Gölitzsch near Merseburg
(Christensen 2004.139).
Arrows
The Apaches used two types of arrow, made either
from reed for close range or full hard wood for long
range (Opler 1996.388). The Apache had learned
about using full-wood arrows from the Navajo (Bas-
so 1971.230). Betsinez could not remember the
name of the bush that produced straight stalks (Bet-
sinez 1959.6), but the Western Apaches used desert
broom (Baccharis sarothroides) (Basso 1971.229).
The Chiricahuas could also use Apache plume, mul-
berry, mountain mahogany or Fendlera rupicola (Op-
ler 1996.388). The shorter full wood arrows were
heavier than the longer reed ones. The wood was
cut into appropriate lengths, peeled of bark and
scraped and dried for three to five days. The sticks
were straightened by pressing between the teeth or
against a heated rock (Opler 1996.388).
Reed arrows were actually composite. The front part
with the point and the end part for the flight feath-
ers and nock were made of hardwood, while the
middle section was reed cane. The best reed was col-
lected along the San Carlos River, but also deeper in-
side Mexico (Basso 1971.227). Reeds were collected
green and fresh. They were straightened between
the teeth and then allowed approximately two weeks
to dry. Straightening could also be achieved by heat-
ing, and then rubbing the reed with a piece of heat-
ed discarded pottery (Opler 1996.390). The arrow
shafts were smoothed by a stone arrow-smoother
(Basso 1971.229), which was made of a porous,
whitish soft stone collected in Bear Canyon. A notch
was made in the stone and the arrow was run
through the groove when the stone was heated.
Reed arrows were typically about 75cm long, while
those made entirely of hard wood were about 5–
10cm shorter (Opler 1996.390). The size of the ar-
rows and bow was further adapted to the needs of
the particular user.
The flight feathers could be from different birds,
although turkey feathers proved the most durable
(Opler 1996.388). The feathers of eagles and small-
er birds of prey such as the red-tailed hawk (Basso
1971.227) were also highly prized, but were hard to
collect (Barrett 1996.68; Betsinez 1959.6). Tail or
wing feathers were split in half and fixed to the
shaft with sinew; sometimes an adhesive made of
pitch was also applied. Usually, one feather was
fixed first and the other two were added one by
one. The parts of the flight had to be of different
feathers, and had to have the same direction of ro-
tation (Basso 1971.230; Opler 1996.389). The length
of particular feathers was 12–15cm (Basso 1971.
388). For a child’s arrow or an arrow intended for
hunting birds and small animals, a single long feath-
er could have been wound spirally around the shaft
for a length of about 20cm (Opler 1996.390).
The Western Apaches painted the last part of the
arrow in red-black using hematite and charcoal as
pigments; the dye was fixed with piñon pitch (Bas-
so 1971.230). The painted shafts were polished with
yucca leaves. The Chiricahuas decorated their arrows
with black, blue or red stripes. A narrow band of
green or red was traced on the upper part of the
shaft that was to hold the feathers. Arrows were
sometimes fluted with a sharp stone, or a spiral was
made in them (Opler 1996.388–389). The fluted ar-
row might have resembled a crawling snake (Opler
1996.286). The style of the arrow generally indicat-
ed which tribe had made it (Basso 1971.229).
The front shaft of the reed arrow was about 10cm
long, approximately half of which was in the reed
and half protruding outside (Opler 1996.390). The
shaft was set into the cane with pitch. The shaft held
the arrow point, which could be of four types: stone,
steel, sharpened wooden tip and wooden shaft with
a four crosspiece rig (Basso 1971.230–231). The
points filed from iron or steel were used in the 19th
Archery by the Apaches – implications of using the bow and arrow in hunter-gatherer communities
521
century when flint points were already a thing of
the distant past (Betsinez 1959.5). Hardwood tips
were additionally hardened by fire (Opler 1996.
389–390). The Western Apaches used hardwood
points equipped with a crosspiece for hunting birds
(Basso 1971.231). The tip of the arrow was split
with two notches, the crosspiece was inserted and
the tip was bound firmly together with a sinew.
Flint points are interesting, of course, as they may
resemble Stone Age examples from Europe. Unfortu-
nately, the Apaches were not highly skilled at stone
knapping. They usually sought flint from around an-
cient ruins (Basso 1971.231) and often used ready-
made points (Opler 1996.389). They knew the arte-
facts belonged to an older people, and thought that
some of the elaborate tools they found belonged to
the mythical ‘Thunder People’ (Opler 1996.389) or
that animals and birds had made them and used
them to fight each other (see Barrett 1996.49; Opler
1994.23–27). There is a record of Chiricahuas split-
ting slate and using the most likely pieces for arrow-
heads (Opler 1996.389). It seems the Western Apa-
ches were more advanced at working flint. They are
documented as having held the flint with a piece of
buckskin in the palm and using deer horn about
13cm long to flake and shape the stone into its de-
sired form (Basso 1971.231). An interesting record
of splitting flint by heating it in fire and cooling it
in water can be found in the memoirs of Herman
Lehman (1993.93). However, this description has to
be taken with reservations, because in the opinion
of certain Comanches and Kiowas, the story is false
(Robinson 2000.63).
To mount the arrowhead, the shaft was split or a
notch cut; the point was inserted and the shaft was
firmly tied on with moistened sinew (Opler 1996.
389, Basso 1971.231).
Quivers and bow covers
Quivers were made of various skins and usually
contained thirty to forty arrows (Basso 1971.234).
Mountain lion skin was the most popular (Barrett
1996.68; Basso 1971.234; Opler 1996.391), while
deer, wolf, wildcat, antelope and otter were used, as
well as domestic animal skins in later periods. The
tail and legs were left to hang, and red flannel and
brass studs could be added as additional decoration.
The quiver was worn on the back, with the strap
passing over the right shoulder and under the left
arm, making it easy to draw an arrow with the right
hand (Basso 1971.234). The Chiricahuas reversed
the quiver in combat, so that the feathers came un-
der the left armpit. Sometimes the quiver was held
over the chest (Opler 1996.341).
The Western Apaches used plain quivers, while the
Chiricahuas preferred quivers combined with bow
cases. To make the quiver and bow cover, women
usually prepared the skin, men cut out the desired
pieces, while sewing was again a woman’s task (Op-
fer 1996.380, 391).
Poisoned arrows
The use of poison is seldom mentioned with respect
to Apache warfare or hunting. The use of arrows
soaked in a mixture of rattlesnake blood, prickly
pear pulp and charcoal from a tree struck by light-
ing is known for the Navajo (Hampton Sides 2007.
3), but these have a more ritual or symbolic mean-
ing. Lehman also mentions that the quiver contained
a few arrows poisoned with rattlesnake venom to
be used in battle (Lehman 1993.94).
The Chiricahuas occasionally used poisened arrows
(Opler 1996.340). The poison was prepared from
deer blood and plants believed to be poisonous. The
mixture was left to rot and then put on the arrow
points; variations of recipes including animal parts
and different plants also existed. If somebody was
hit with a poisoned arrow, he was expected to turn
black.
The Western Apaches seemed to have been more
inclined to use poison in combat and hunting (Bas-
so 1971.232–234). The poison was made of deer
spleen. First it was dried, then mixed with the
ground roots or stalk of nettles and some other
plants with a burning taste. The ingredients were
put in a sack of deer intestine and hung on a tree
for about three to five days. The liquid was then
painted on the arrow tips. The dried poison was also
efficient if ground and mixed with spit. A deer pierc-
ed or merely scratched with a poisoned arrow could
run about seven hundred meters before falling, but
the poison spoiled only the meat near the wound.
The experiment
Archaeological examples
Remnants of reed arrows have been discovered in
several caves, such as the Fort Stanton Cave. The re-
construction of reed arrows can be seen in the mu-
seums in San Carlos, Mescalero and Fort Stanton. On
territory once settled by the Apaches, the author
?iga {mit
522
noted several traces of the flint industry at the
northern entrance to the Ojo Caliente canyon (Win-
ston, New Mexico) and at Turkey Creek (Fort Apa-
che, Arizona). Several pieces were triangular in form,
but it is impossible to guess if they were indeed used
as arrow heads. The only fragment where the iden-
tification of an arrow head is not problematic is
shown in Fig. 3. It is the lower part of a triangular
head made of white flint without a stem, but with
a clear notch for mounting into the split shaft. It is
11.6mm long and weighs 0.36g, so we may extrapo-
late its full length to about 25mm and weight to
about 0.6g.
Making an arrow
The reed arrow was manufactured from material
collected in Slovenia, so in certain aspects it departs
from historical examples. We also used modern
thread and modern synthetic rubber or cellulose
acetate glue for fixing; however, we do not believe
these additions greatly affect its ballistic properties.
The wooden parts were made from straight shoots
of dogwood, which has compact wood. For the
point, a triangular flake of flint weighing 0.4g was
inserted into the split in the fore shaft, fixed with
cotton thread and glued with synthetic rubber. The
wood was split by a saw so as to prevent the shaft
from cracking below the point. The shaft was made
from a dry reed. After removing the dry leaves, we
noticed the importance of properly smoothing the
leaf nodes on the stem. We suppose historical arrow-
smoothers were mainly used for this type of work.
We made a proper notch and fixed three feathers
at the end of the shaft, firstly with thread around
the front and end parts; finally, we also applied a
small amount of glue below the feather’s pulp. The
feathers were taken from small birds the size of a
pigeon or magpie, which are softer than turkey feath-
ers. Fixing the feathers was the most delicate and
time-consuming part of the work (see also Odar
2011.453). The end parts of the reed stem were re-
inforced with cotton fibre and glue. For the final as-
sembly of the three main parts, the sockets were
gently wetted with cellulose acetate glue. The linear
dimensions and weights of particular parts are shown
in Table 1; details can be seen in Fig.4. Our final re-
sult was an arrow 70.7cm long, but with a remark-
ably low weight of 9.7g.
The bow
We made the bow by following the Apache tradition
approximately, as we wanted to produce only a
wooden bow with similar properties to suit our bal-
listic experiments. Most of the mulberry trees in Slo-
venia are old trees with massive trunks and young
branches too thin for bow making. We were able to
cut a branch from a yew with the appropriate length
and thickness for shaping into a bow. One part was
rasped in a limb, applying distinctive tapering, while
the other part had to be tapered on the inner side
only. The shaped bow was dried for three years be-
fore being used. A disadvantage of using such a
branch was found to be in the curvature of the bow:
it remained significantly curved even when unstrung.
The string was made of hemp cord 1mm thick,
which was doubled by twisting and then treated
with bee wax. When strung, the bow was 105.5cm
long (Fig. 5). The weight of a full pull for a 70cm ar-
rowhead was about 18kg, but we usually pulled the
bow at a weight of 16kg. We also measured the pull
force as a function of the arrow draw distance and
estimated that one third of the bow’s potential ener-
gy is released as kinetic energy in the arrow.
Shooting experiments
As a first attempt, we tried to determine arrow veloc-
ity with an apparatus used to measure the velocity
of bullets, but the arrow’s velocity was evidently too
low to be detected, or the length of the arrow did
not produce a short cut-off signal. Therefore, we
tried to deduce the take-off velocity from the arrow
range when the bow was fired at different angles, si-
multaneously measuring the flight time. The parame-
ters of the ballistic trajectory can then be determin-
ed by means of a model calculation, assuming that
Fig. 3. Fragment of a flint arrow head (Turkey
Creek) with supposed reconstruction.
Tab. 1. Dimensions and weights of reed arrow
parts.
Length\with socket Mass
(cm) (g)
Front shaft with flint tip 11.7\17.6 3.54
Reed middle shaft 46.3 3.51
Feathered end shaft 12.7\15.5 2.65
Total 70.7 9.70
Archery by the Apaches – implications of using the bow and arrow in hunter-gatherer communities
523
the air drag force varies with
the square of the projectile’s
velocity (Laharnar 2011.77–
80).
Flight time was measured
with a stop watch held by the
archer in the left hand toge-
ther with the bow and start-
ed at the same time as the ar-
row was released. The watch
was stopped as soon as the
archer saw the arrow touch
the ground. Manipulating the stop watch prevent-
ed the archer from achieving maximum draw, so
we may estimate that the draw weight of the bow
was approx. 16kg. The angle of the arrow with re-
spect to ground was also observed and compared
with the calculated angle. Range was measured with
a steel measuring tape with accuracy better than
0.2m. The experiments were performed at 300m
above sea level at temperatures around 20°C. The
reduced air pressure on the day of the experiment
was 1015mbar. Subjectively, the feeling of shoot-
ing the reed arrow was good. It was possible to per-
ceive that the arrow left the bow at a significant
speed, but drag slowed it down noticeably after
about 20m. The flight was stable and did not wob-
ble, in spite of the shaft not being completely straight.
The results of the experiment are presented in the
first three columns of Table 2.
By trial and error, we then looked for the ballistic
parameters that matched our measured ranges and
flight times. The variable parameters were take-off
velocity, take-off angle and effective arrow cross
section (the physical cross section multiplied by
the drag coefficient). Three independent parame-
ters are too many to determine a unique ballistic tra-
jectory, so another principle was selected to discov-
er the effective arrow cross section, which should be
the same for all trajectories. As a starting guess, we
used data from Jürgen Junkmanns (2001.57, 58). A
yew bow with 28kg pull yielded a take-off velocity
of about 50m/s for a flint-tipped arrow with a weight
of 30g, which resulted in a range of 120m. The sim-
ulation gave the effective cross section as 3.5cm2.
We then calculated five sets of take-off velocities for
an effective cross section varying between 3.25 and
4.25cm2 and determined their mean and standard
deviations. The realistic cross section corresponded
to the values where the spread of take-off veloci-
ties was minimal. This principle is based on the sup-
position that all arrows were shot at approximately
the same velocities at different angles. The first two
shots in Table 2 were omitted from the calculation,
as the archer was still learning to use
the bow whilst holding the stop-
watch. By plotting the chi-squared
rather than standard deviation for
the calculated take-off velocities (Fig.
6), we found the effective cross sec-
tion of our arrow was 3.61±0.5cm2.
This value was then used to deter-
mine the ballistic parameters in the
second three columns of Table 2.
Discussion of shooting experi-
ments
The ranges in Table 2 appear rather
low, but one should remember that
they were produced for an extreme-
ly light arrow. The ranges and times
of flight imply take-off velocities of
about 40m/s, which is also not high.
Fig. 4. The reed arrow used for the ballistic experiments.
Fig. 5. The bow used in the experiments
with its draw characteristics.
?iga {mit
524
However, even if the take-
off velocities were 60m/s
and arrows were fired at
2000m a. s. l., the maxi-
mum range would not ex-
ceed 83m. For a take-off ve-
locity of 45m/s, the arrow’s
kinetic energy is 10J, which
is one fifth of the energy of
22LR bullets, and only ten
times more than the energy
of an air-soft pellet. At a di-
stance of 10m, this energy
is reduced to about 6.2J, and at a distance of 15m
it drops to 5J, half of its initial value.
For comparison, a 30g arrow with a take-off velocity
of 50m/s (having thus initial kinetic energy of 37.5J)
would strike a target 100m away with an energy of
11J. We can conclude that the effective range of reed
arrows was well below 20m. The main advantages of
such arrows were that they could reach a reason-
able velocity from a bow that was not very strong (as
in our experimental case). Using a light bow further
enabled the archer to fire more arrows in a short
time, which might be essential in a dense fighting
melee, and eased aiming, as the trajectory of the light
arrow for the first 20m is almost straight (Fig. 7).
Conclusion
The use of reed arrows by the Apaches provided an
efficient weapon for close-range hunting and com-
bat. This selection of reed as arrow material is not
unique and may not have developed strictly for func-
tional reasons, but because material which is found
in the woodland and deserts of the American south-
west is less generous with long straight sprouts than
bushes in Europe. Nevertheless, historical Apaches
used the bow as their principal long-range weapon
until the middle of the 19th century and developed
a mastery that can serve as an example for current
prehistoric models. As the use of light arrows in
combat favoured close-range shooting, the rock
painting in Cueva del Roure showing Neolithic ar-
chers in a short-range fight may indeed represent a
realistic case.
range time estim. angle to calculated take-off calc. take-off calc. angle to
(m) (s) ground (deg.) velocity (m\s) angle (deg.) ground (deg.)
44.5 2.8 – 33.5 32 50.6
48.5 3.4 60 36.1 39 59.9 
47.5 4.0 60 37.6 49 69.1
35.5 4.9 75 39.2 66 79.2
49.5 3.8 60 38.0 45 66.2 
51.5 3.5 60 38.7 39 61.2
Tab. 2. Results of shooting experiments (first three columns from left) and
deduced take-off velocity, take-off angle and angle at which the arrow hit
the ground for an effective arrow cross section of 3.61cm2. Air density was
taken as 1.16mg/cm3.
Fig. 6. Determination of effective arrow cross sec-
tion as a value where the calculated take-off velo-
cities have minimal spread. The fitted curve is
fourth order polynomial.
Fig. 7. Numerical simulation of the ballistic tra-
jectories for the longest (1) and highest shot (2) in
the experiment and almost flat trajectory for close-
range shooting to a distance of 20m (3).
Archery by the Apaches – implications of using the bow and arrow in hunter-gatherer communities
525
References
Ball E., Kaywaykla J. 2003. In the Days of Victorio: Re-
collections of a Warm Springs Apache (1st edition in
1970). The University of Arizona Press. Tucson.
Ball E., Henn N. and Sánchez L. A. 1988. Indeh: An Apache
Odyssey. University of Oklahoma Press. Norman, London.
Barrett S. M. 1996. Geronimo, His Own Story. As Told
to S. M. Barrett. A Meridian Book. Penguin Books. New
York.
Basso K. H. (ed.) 1971. Western Apache Raiding and
Warfare. From the notes of Grenville Goodwin. The Uni-
versity of Arizona Press. Tucson.
Betsinez J., Nye W. S. 1959. I Fought with Geronimo. Uni-
versity of Nebraska Press. Lincoln, London.
Bonsall C. 2008. The Mesolithic of the Iron Gates. In G.
Bailey, P. Spikins (eds.), Mesolithic Europe. Cambridge
University Press. Cambridge: 238–452.
Bourke J. G. 1958. An Apache Campaign in the Sierra
Madre. An account of the expedition in pursuit of the
hostile Chiricahua Apaches in the spring of 1883 (1st
edition in 1886). Charles Scribner’s Sons. New York 1958.
Chatters J. C. 2014. Wild-type colonizers and high levels
of violence among Paleoamericans. In M. W. Allen, T. L.
Jones (eds.), Violence and warfare among hunter-gath-
erers. Left coast press. Walnut Creek: 70– 96.
Christensen J. 2004. Warfare in the European Neolithic.
Acta Archaeologica 75: 129–156.
Davis B. 1976. The Truth about Geronimo (Fascimile of
1929 Edition). University of Nebraska Press. Lincoln, Lon-
don.
Erdal Y. S., Dilek Erdal Ö. D. 2012. Organized violence in
Anatolia: A retrospective research on the injuries from
the Neolithic to Early Bronze Age. International Journal
of Paleopathology 2: 78–92.
Hutton Estabrook V. 2014. Violence and warfare in the
European Mesolithic and Paleolithic. In M. W. Allen, T. L.
Jones (eds.), Violence and warfare among hunter-gath-
erers. Left coast press. Walnut Creek: 49–69.
Junkmanns J. 2001. Pfeil und Bogen. Herstellung und
Gebrauch in der Jungsteinzeit. Museum Schwab. Biel.
Kramer C. 1979. Introduction. In C. Kramer (ed.), Ethno-
archaeology. Implications of Ethnography for Archaeo-
logy. Columbia University Press. New York: 1–20.
Laharnar B., πmit Ω. and Ravbar I. 2011. Sling and sling-
shots – An example of experimental archaeology and ap-
plication of statistical methods in archaeology. Arheo 28:
73–86.
LeBlanc S. A. 2014. Forager warfare and our evolutionary
past. In M. W. Allen, T. L. Jones (eds.), Violence and war-
fare among hunter-gatherers. Left coast press. Walnut
Creek: 26–46.
Lehmann H. 1993. Nine Years among the Indians (1st
edition in 1927). University of New Mexico Press. Albu-
querque.
Odar B. 2011. Archers at Poto≠ka zijalka. Arheolo∏ki vest-
nik 62: 433–456.
Opler M. E. 1994. Myths and Tales of the Chiricahua
Apache Indians (1st edition in 1942). University of Ne-
braska Press. Lincoln, London.
1996. An Apache life-way. The economic, social, and
religious institutions of the Chiricahua Indians. Uni-
versity of Chicago Press. Chicago.
Robinson S. 2000. Apache Voices. Their Stories as Told
to Eve Ball. University of New Mexico Press. Albuquerque.
Roksandi≤ M. 2006a. Violence in the Mesolithic. Docu-
menta Praehistorica 33: 165–182.
2006b. Interpersonal violence at Lepenski Vir, Mesoli-
thic/Neolithic Complex of the Iron Gates gorge (Serbia-
Romania). American Journal of Physical Anthropology
129: 339–348.
Sides H. 2007. Blood and Thunder, The epic story of Kit
Carson and the conquest of the American West. Anchor
Books. New York.
Sonnichsen C. L. 1958. The Mescalero Apaches. Univer-
sity of Oklahoma Press. Norman.
Sweeney E. R. 1991. Cochise, Chiricahua Apache Chief.
University of Oklahoma Press. Norman.
2010. From Cochise to Geronimo: The Chiricahua
Apaches, 1874–1886. University of Oklahoma Press.
Norman.
Thrapp D. L. 1974. Victorio and the Mimbres Apaches.
University of Oklahoma Press. Norman.
Utley R. M. 1977. A Clash of Cultures, Fort Bowie and the
Chiricahua Apaches. National Park Service. Washington.
2012. Geronimo. Yale University Press. New Haven,
London.
.