A. L. (Alfred Louis) Kroeber.

Basket designs of the Indians of northwestern California online

. (page 1 of 5)
Online LibraryA. L. (Alfred Louis) KroeberBasket designs of the Indians of northwestern California → online text (page 1 of 5)
Font size
QR-code for this ebook


Vol. 2 No. 4







The publications issued from the Department of Anthropology of the
University of California are sent in exchange for the publications of
anthropological societies and museums, for journals devoted to general
anthropology or to archaeology and ethnology, and for specimens
contributed to the museum collections of the Department. They are
also for sale at the prices stated, which include postage or express
charges. They consist of three series of octavo volumes, a series of
quarto memoirs, and occasional special volumes.


Vol. 1. The Tebtunis Papyri, Part I. Edited by Bernard P. Grenfell, Arthur
S. Hunt, and J. Gilbart Smyly. Pages 690, Plates 9, 1903
Price, $16.00

Vol. 2. The Tebtunis Papyri, Part 2 (in preparation).


Vol. 1. The Hearst Medical Papyrus. Edited by G. A. Reisner and A. M.
Lythgoe (in press).


Vol. 1. No. 1. Life and Culture of the Hupa, by Pliny Earle Goddard.

Pages 88, Plates 30, September, 1 1903 . . . Price, 1.25

No. 2. Hupa Texts, by Pliny Earle Goddard. Pages 290, March,

1904 Price, 3.00

Vol. 2. No. 1. The Exploration of the Potter Creek Cave, by William J.

Sinclair. Pages 27, Plates 14, April, 1904 . . Price, .40
No. 2. The Languages of the Coast of California South of San

Francisco, by A. L. Kroeber. Pages 72, June, 1904. Price, .60
No. 3. Types of Indian Culture in California, by A. L. Kroeber.

Pages 22, June, 1904 Price, .25

No. 4. Basket Designs of the Indians of Northwestern California,

by A. L. Kroeber. Pages 60, Plates 7, January, 1905. Price, .75
Vol. 3. The Morphology of the Hupa Language, by Pliny Earle Goddard

(in press).


Vol. I. Explorations in Peru, by Max Uhle (in preparation).
No. 1. The Ruins of Moche.
No. 2. Huamachuco, Chincha, lea.
No. 3. The Inca Buildings of the Valley of Pisco.


The Book of the Life of the Ancient Mexicans, containing an account of their
rites and superstitions; an anonymous Hispano-American manuscript
preserved in the Biblioteca Nazionale Centrale, Florence, Italy. Repro-
duced in fac-simile, with introduction, translation, and commentary,
by Zelia Nuttall.

Part I. Preface, Introduction, and 80 Fac-simile plates in

colors. 1903.

Part II. Translation and Commentary. (In press).
Price for the two parts $25.00

Address orders for the above to the University Press, Berkeley,
California. Exchanges to be addressed to the Department of Anthro-
pology, University of California, Berkeley, California.

A. L. KROEBER, Secretary. F. W. PUTNAM, Director.


VOL. 2, PL, 15.

"^t^ **i^^^MiNktjj^*'

Caps. Yurok.


VOL.2 No .





The Indians of extreme northwestern California, while show-
ing many similarities to the other tribes of California, and
some approximation to those of the north Pacific coast, are in
many ways peculiar in their culture. The territory occupied by
this group of tribes is very limited, comprising only Humboldt
and Del Norte and small parts of Trinity and Siskiyou counties.
Their specialized culture is found in its most highly developed
form among the tribes of the lower Klamath and Trinity rivers :
the Yurok, Karok, and Hupa. The Hupa belong to one of the
California groups of the great Athabascan linguistic stock. The
Yurok and Karok are small isolated linguistic stocks. The three
languages are as radically different in phonetics as they are
totally unrelated in vocabulary. The three tribes live in close
contact, with more or less intercourse and generally friendly
relations. In their culture they are remarkably alike.

The names of the basket designs described in this paper were
obtained from Indians of the three tribes during 1900, 1901, and
1902. The most extensive investigations were made among the
Yurok. This accounts for the larger number of designs obtained
among this tribe. The Yurok designs described are taken from
nearly a hundred baskets. The majority of these are now in the
Museum of the Anthropological Department of the University
of California. A number of baskets, and the names of their
designs, were collected in 1900 for the California Academy of
Sciences. Through the courtesy of the officers of the Academy
this material is used in the present paper. Information was

AM. ABCH. ETB. 2. 9.

106 University of California Publications. [AM. ARCH. ETH.

obtained among the Yurok as to the designs of a greater number
of baskets than were actually collected, the total number reach-
ing several hundred. The more common design names are
exceedingly frequent among the northwestern tribes, and, while
exact duplications of designs ordinarily do not occur, yet many
of the variations are so slight that it was often thought unneces-
sary to insure their preservation by purchase of the specimen.
All baskets having characteristic designs but uncommon design-
names were secured for the Museum of the Department. This
selection gives the Yurok design names described an appearance
of somewhat greater variety than they actually possess. Prob-
ably the fifteen most common design names constitute all but a
very few per cent of the total number. Among the Karok and
Hupa all baskets were secured about which information was
obtained as to the design. The number of such Karok baskets
is about fifty, and of Hupa twenty-five.

It was found necessary to get the names of the designs in
the native language, as many of the words are not names of ani-
mals or objects, but geometrical or descriptive terms not trans-
latable by the Indians. 1


The basketry of northwestern California is characterized by
"circular open baskets somewhat rounded at the bottom and
generally of no very great depth, "and by women's caps, which
are shallower than the basketry caps worn in other parts of Cali-
fornia. Large baskets serving for the storage of food are propor-
tionally of deeper shape than the smaller baskets used for cook-
ing and eating. Conical baskets are used for gathering seeds,
and flat circular baskets for trays, plates, and meal sifters. The
acorn mortar consists of a basket hopper of the type used by the
Porno. Conical carrying baskets, baby baskets, plates, and some
trinket baskets are made in open work. The various kinds and-

1 The following characters have been used : c = sh, x = spirant of
k = kh, q = velar k, L = palatal or lateral 1, n = ng ; a a as in father ;
a a as in bad ; a = English aw ; e and 6 = long open e and o ; A, E,
i, o, u, = obscure vowels. Yurok r has the peculiar quality of American r
in an exaggerated degree. Karok r is clear and trilled. Yurok v is bilab-
ial, having nearly the the sound of w, and its g is always a spirant
g' = gh.

VOL. 2] Kroeber. Basket Designs of N.W. California. 107

shapes of baskets can be seen in the accompanying plates 15 to
21, and in plates 20 to 27 published in the first volume of the
present series of University of California publications.

Yurok names for baskets are: waxpeya, cap, if brown (Plate
15, figures 7, 8) ; aqa', cap, if the ground is covered with over-
laying (Plate 15, figures 1 to 6) ; he'kwuts, small basket for acorn
mush, especially for eating (Plate 16, figure 3, and figure 6,
unfinished) ; muri'p, large basket for acorn mush, used for cook-
ing (Plate 16, figures 4, 5; he'kwuts and muri'p are called by
the Karok asip : Plate 20, figures 4, 5, 6, 8) ; perxtse'kuc, a basket
higher than he'kwuts, used for keeping small objects (Plate
17, figures 4, 5, 6; Karok cipnuk, Plate 20, figure 3) ; rumi'tsek,
an openwork trinket basket (Plate 19, figure 5, usual form; fig-
ure 6, unusual) ; qewa'i, conical burden basket of openwork (see
P. E. Goddard, Life and Culture of the Hupa, University of
California Publications, American Archaeology and Ethnology,
I, Plate 22, figure 1) ; terre'ks, conical basket for gathering seeds
(Goddard, op. cit., Plate 22, figure 2, of Yurok provenience) ;
paaxte'kwc, basket for storing food, especially acorns, much like
perxtse'kuc but much larger (Goddard, Plate 23, figure 1, a
Yurok specimen) ; meixtso', storage basket similar in shape, but
made altogether of hazel, without overlaying or patterns ; poixko',
large flat tray for acorn meal (Goddard, Plate 24, figure* 2) ;
poixtse'kuc, small tray for seeds used as food (Plate 19, figures
1, 2), also small, flat, conical dipper for acorn mush (Plate 19,
figure 3, a Karok specimen) ; wetsane'p, meal sifter, flat without
appreciable curvature (Plate 18, figure 2); laxp'ceu, openwork
plates for eating salmon (Plate 18, figures 1, 3; Goddard, Plate
21, figure 2, a Yurok specimen) ; meco'liL, larger openwork
plates on which salmon is laid; upe'kwanu, mortar hopper (God-
dard, Plate 24, figure 1, Yurok) ; qeme'u, also called haxku'm
uperxtse'kuc, "tobacco its storage-basket," tobacco basket, often
with a lid, and similar to the perxtse'kuc, though generally
smaller (Plate 17, figures 1, 3, 5, 7, Plate 19, figure 4) ; uqem'-
te'm, said to have been a large form of perxtse'kuc with a small
opening and a lid, used for storage of valuable property ; ego'or,
an approximately cylindrical basket used in the jumping dance,
made of a rectangular sheet bent into shape of a cylinder slit

108 University of California Publications. [ AM - ARCH. ETH.

along the top (Plate 18, figure 4). A Hupa baby basket and
seedbeater are shown in Goddard's Plate 21, figure 1, and Plate
23, figure 2. The aqa', perxtse'kuc, terre'ks, paaxte'kwc, poixko',
poixtse'kuc, wetsane'p, qeme'u, uqem'te'm, and ego'or are gene-
rally overlaid with white; the waxpeya, he'kwuts, muri'p, upe'-
kwanu, and sometimes the poixtse'kuc, are mostly in unover-
laid brown, but usually with a pattern in overlaying; the rumi'-
tsek, qewa'i, laxp'ceu, meco'liL are in openwork.


The basket materials of this region and their employment
have recently been given full treatment in Dr. P. E. Goddard's
Life and Culture of the Hupa, 1 and on a less localized basis by
F. V. Coville in Professor 0. T. Mason's Aboriginal American
Basketry. 2

According to information obtained from the Yurok, the
warp of their basketry regularly consists of hazel twigs. The
woof is made of strands from roots of sugar pine and near the
coast of spruce. Redwood and willow roots are inferior but
used. Willow seems to be usual for the woof in beginning a

While these root fibres give a colorless gray, deepening with
age to a not unpleasant brown, designs and sometimes the entire
ground color are produced by overlaying in other materials.
The most important of these is the widely used and well known
lustrous whitish grass xerophyllum tenax. In baskets for ordi-
nary use the designs are worked in this white on the darker
ground of root-fibre woof. In ornamental baskets the ground is
overlaid with this material, and the patterns are black, red, and
occasionally yellow. For black the outside of stems of a species
of maidenhair fern, adiantum, are used; for red, alder-dyed
fibres of a large woodwardia fern. The stems of this fern are
bruised by beating, and two flat fibres extracted from each.
These are usually dyed by being passed through the mouth after
alder bark has been chewed. Yellow is produced by dyeing with

1 Univ. Gal. Publ., Am. Arch. Ethn., I, 38 seq., 1903.

2 Rep. U. S. Nat. Mus. 1902, 199 seq., 1904.

VOL. 2] Kroeber. Basket Designs of N.W. California. 109

a lichen, the widely used evernia vulpina. Porcupine quills dyed
yellow are rarely used. 1

Besides red and yellow, black dyeing is occasionally prac-
ticed by burial of materials in mud. Part of the hazel twigs
for the warp of openwork plate baskets are sometimes treated
in this way; and rarely the woodwardia fibre for the woof of
other baskets.

Of the three colors used on a white ground, black most fre-
quently stands alone. Red is usually accompanied by at least
a certain amount of black ornamentation, such as lines or edg-
ing. Yellow does not seem to be used without accompanying red
or black, usually the latter. Occasionally the three colors are
used in combination on a white ground, but although pleasing
if skilfully carried out this is uncommon. Sometimes areas of
unoverlaid brown are left in colored baskets and employed in
design effects. The only baskets with unoverlaid ground whose
patterns sometimes contain black or red in addition to white, are
hats, even the plainest of which, as is only natural, show more
ornamentation than is usual in baskets for household purposes.

A somewhat greater proportion of red to black designs is
found among the Karok than among the Yurok or Hupa, due
possibly to greater scarcity of the maidenhair fern furnishing



In regard to technique, the fundamental feature of the bas-
ketry of northwestern California is that twining is the only
method followed. Coiled weaves of any kind, except as a border
finish, are unknown. This statement can be made without quali-
fication, and all coiled baskets attributed to this region are of
erroneous provenience or obtained by the northwestern Indians
from more southerly tribes.

To all intents these Indians practice only one weave, the
simple twining with two strands. This is used for the finest
hats, for the largest and coarsest storage baskets, for cooking
baskets, and for openwork plates, cradles, and carrying baskets.

1 Yurok names of basket materials and dyes: hali'L, hazel; paxkwo',
willow; waxpe'u, sugar pine; qiL, redwood; teiwolite'po, spruce; haamo',
xerophyllum tenax; rego'o, maidenhair fern; paap, woodwardia fern;
were'regets, alder; mece'n, evernia lichen.

110 University of California Publications. [ AM - ARCH. ETH.

Though two-strand twining is very close to wickerwork, differ-
ing from it only in that the two strands cross after each warp
is passed, instead of continuing parallel, these tribes do not seem
to practice wickerwork.

Three-strand twining is well known in this region and fre-
quent in use, but apparently no baskets are made completely in
this weave. Almost all baskets begin in this weave ; the majority
have one or more courses of it where the bottom begins to turn,
and again near the top; and occasionally a basket is finished in
it. The specific technique seems to be simple three-strand twin-
ing, not three-strand braiding. Each woof strand passes over
two warp rods on the outer or pattern side of the basket, over
one on the inside.

There is one basket in the collections of the Department of
Anthropology from this region in which the two strands of the
woof cover two rods of the warp at a time, while in the following
course they take these rods so as to alternate with the previous
one. This is the weave that has been called diagonal twining.
The basket is shown in Plate 17. At its origin it shows the usual
three-strand twining. While the alternate or diagonal weave
has been praised by Mason and Purdy as more susceptible of
developed decoration than ordinary twining, this basket is unor-
namented except by two plain bands. This poverty of decora-
tion is perhaps due to the fact that the ornamentation is pro-
duced by covering of the woof instead of by the woof itself.
One or two other baskets found are made in this weave for a
number of courses near their origin.

In two-strand twining the woof strands are usually more or
less flat, and are not twisted, the same side being turned toward
the outside of the basket continuously, whether overlaid or not.

The only usual modification of two strand twined weaving
is a multiple warp. This is common for the bottom of large
storage baskets, and is usually accompanied by a certain degree
of openness of woof. After the turn from the horizontal bottom
has been made and the sides of the basket started on their upward
course, the additional warp sticks taper out and are dropped
and the weave is continued on the main stick of each group.
Sometimes a group is so divided as to result in two single warp

VOL. 2] Kroeber. Basket Designs of N.W. California. m

Crossing of the warp sometimes occurs in openwork, most
often for one course just below the border, occasionally near the

Strengthening by means of a rod enclosed in the twining
is common. This forms the first step toward lattice twining or
the ti weave, a superimposition of coiling on twining. Mortar
baskets are strengthened by several stout rods; storage baskets
frequently show one or two near top or bottom ; and occasionally
a rod is used as a finish. The great majority of cooking baskets
have two strands, apparently of root, laid around the outside
near the top of the basket in the region of the typical design
zone, which they serve markedly to define, limit, or divide. It
is probable that their decorative effect is their chief purpose;
being pliable, they do not stiffen the basket appreciably, and
being held only by the twining of the overlaying material the
body of the woof being usually completely lacking in the two
courses on which the strands are laid they can scarcely be a
source of strength.

Ornamentation almost without exception is produced by over-
laying or false embroidery, and not by the use of colored or dyed
woof materials. The method of overlaying differs from that of
the Tlinkit and Thompson Indians, two strands being employed
instead of one. Among the Tlinkit "the decorative element,
instead of taking its turn to pass behind the warp, remains on
the outside and makes a wrap about the strand that happens to
be there." The Thompson Indians follow a method of "passing
a strip of ... material entirely around the twining each time,
showing the figure on the inside." 1 In northwestern California
each of the two woof strands is faced as it were, in the process
of weaving, with a strand of overlaying material toward the out-
side of the basket. This facing follows the woof-strand behind
the warp, and together with it twines with the other woof -strand
and its facing. As the overlaying always faces the outside of
the basket, and not the outside of the twining, each strand of
it is half the time between warp and woof and invisible, and
the decoration does not show on the inside of the basket except
casually between turns and plies especially in coarser baskets.
1 Mason, Aborig. Amer. Baaketry, Kep. U. 8. Nat. Mus. 1902, 309.

112 University of California Publications. [ AM - ARCH. ETH.

Fine hats are nearly as completely free from trace of over-
laying inside as is Tlinkit work. The two overlaying strands
follow the woof strands to the edge of the design-figure, where
they are broken off on the inside of the basket, and the woof
continues on its course alone, or overlaid by strands of a differ-
ent color, until the next figure is reached. Occasionally, where
this intervening space between designs is not great, especially
where there is a small recurrent design, the overlaying is not
broken off, but brought to the rear of the woof, so as to be invis-
ible from the front, and carried along to the next figure, when
it reappears. Of course it then shows inside the basket while it
is invisible on the outside, but this occasional result seems to be
produced among the northwestern tribes not for its effect but
because in such cases it is preferable to carry on the overlaying
material rather than cut the strands to reinsert them a few
turns, sometimes only two or three, farther on.

It will be seen that this method of overlaying cannot be
"classed technically with three strand twined weaving," as
Professor Mason says of the Tlinkit process, not only because
there is a total of four strands in the woof, but because the opera-
tion is essentially one of two-strand twining with double strands.

In northeastern California, among the northeasternmost
Wintun tribes, on the McCloud river, still another process of
overlaying is practiced. Like the northwestern overlaying, this
is done with two strands, but the overlays form a separate twin-
ing around both warp and woof, which latter they entirely
enclose, never being within its plies as in the northwestern
process. The design thus shows inside the basket as well as out-
side. That the difference in this respect from the northwestern
basket is fundamental, is evidenced by the fact that in the cases
when the design appears on the inside of a northwestern basket
it does so in the intervals of its disappearance from the outside,
the inside and outside figures being the reverse of each other;
whereas in these North Wintun baskets the regular overlaying
appears inside in the same places as outside and forms identical
figures. In the northeastern weaving each strand of overlay is
evidently carried and treated as part of one of the woof strands,
as in the northwestern process, but in passing around each warp


VOL. 2, PL. 16.

I. 12, :t, "), (i. Cooking Iciskrts. Vurok.
Fit:. 4. Cooking liaskct. Knn.k. !.

VOL. 2] Kroeber. Basket Designs of N.W. California. 113

rod it is either given a half-twist to the other side of the strand
that it accompanies, or much more probably the combined woof
and overlay strand is thus half twisted.

This northern Wintun method of overlaying is used also by
the Lutuami or Klamath Lake and Modoc Indians, and perhaps
by the Achomawi, the Pit River Indians.

The overlaying materials in northwestern basketry are never
used without an underlying woof to serve them as body: but
sometimes this woof is itself of the overlaying material, either
with or without another overlay of the same or another material.
Where a pattern is worked consisting of alternate stitches of
overlaid and of undecorated woof, the whole design being merely
one of regularly disposed dots, the woof strand on which the
white overlay is carried is usually if not always itself of this
material, and sometimes of double thickness, in this case making
a woof of three flat white strands twining alternately with one
of a single strand of brown root fibre. The same process is fol-
lowed to produce a design of vertical bars only one stitch wide
and one stitch apart. It is easy to see why the single overlay
in these cases is carried on continuously with its supporting
woof; but the only explanation that seems to account for the
underlying woof itself being of overlay material is a desire to
preserve the two woof strands of the same total thickness, which,
as only one of them is overlaid, would be very difficult if the
same body material were used for both of them. The white
xerophyllum is flat and thin, so that two or three strands of it
about equal in thickness one of the more rounded root fibres
usually forming the woof.

In some baskets almost completly covered with overlay, por-
tions are sometimes entirely without woof except of overlaying
materials. The motive is apparently the desire to avoid addi-
tional strands in the twining, which would detract from fineness
of stitch ; but as different parts of a basket are sometimes incon-
sistently treated, it is difficult in all cases to follow the weaver's
purpose. A Karok basket covered with a solid pattern of contig-
uous red and white isosceles triangles alternately pointing up and
down, lacks for the major part the usual root woof. Where the
pattern in this basket is white, the red material serves as under-

114 University of California Publications. [ AM - ARCH. ETH.

lay, and consequently appears on the inside of the basket in an
identical red figure ; and vice versa. The purpose of this device
is explicable ; owing to a desire to continue the strands of over-
lay unbroken, the usual colorless woof was sacrificed to avoid
carrying a total of six threads, and its place taken by the overlay
temporarily not appearing in the design. The triangles in this
basket are however separated into several bands by horizontal
lines consisting of a single course of black overlaying. In two
of these courses the woof under the black material consists of
red overlay; but in several other courses the woof is the usual
colorless root fibre; and this material is used also for the woof
of one of the adjacent courses forming part of the triangle

1 3 4 5

Online LibraryA. L. (Alfred Louis) KroeberBasket designs of the Indians of northwestern California → online text (page 1 of 5)