CN!i 16 gA "W7. 41 YON 1111,4?,_F` TY EU4 40 W ?4, "NA,17. JR, 4Z Xj v ?19 ok ,l al 46 IFW IWO 1? 51, -?M 7: 4 W, A, 7i, An V?F W-P -4 ?32 II ir? Af x, 4, ti, Py, Ai, iv T llZ. It J, "'A 1AA ' -AAA-I -' ArA :;"A15S-P : .' AGW-M AJPA- -B BNAE, ONAX -PAS: tAB MAIHF PMAP S'AP -AP ' :- USIN UOPA", U M - A P , ; USN M0 - - R , , U WX -.P A B > :ZVIA:-TIONS USED:- VAnthr4opos.le L'AnthrooIgie ; V>- American Anthropologit. American Anthroplogicala Association, Mm. Archi fvr Anthropol6gie, :;American Bthnoogca Society, Pnblicti ous. -- An *opolgiso1*e, (e' lls4haft I Win, litteug : American-Journal -f Physical Antiropolog.I Amenic-an Mueum ofNa,:tHistory- Anthropologa Papers.' Bulftlitln ; Memoirs, Antopologil Series,. ,ewmosirs, Jesup Expedition. Bureau:of A mer,ican.EthoJogy- , ! Bulletins. Cont,ribhl.utiis, to Nr American Ethnology' Col-Cmbia University,Onbi,ffns t A polo FYOU-Lor"e. -FieldMuseun of Nt Hsory. ' Publiciations, A pological Serie. 1nternationaleg Ari v fu E tnogTaphUe.; International'Oi.grss of Aeats (ates Reudu, Proeei ngs). International Journalof Aerican I aingu.stb s. Journa l,of America l-k-Lore- Jourual of the ZRoyal Anthopical Institute. Museum of AithA ecaIna eye- Foundation-; - : Contriblutins. E- - - Xndlan, w'Notes. :. Indian, Note andlYongrb. , PeabodyIiUseu bf(d Un.ivetity)- Memoirs- Repors i & t * 0 t e ;t' O f - PubBlp Museum (ofthie ity) o,f Mlaukee, Bllein..; '''';--9 t Soci6t6d Am6r.i&:a,,nist' d Paris, 1ournaL SmithsonanInstituion-: * Annual Reorts.-..- .-. .*--.- < ; "Cob to- owlee.4i Mis0ellneu Colletiions.+<;:< trv; ' ,Uuiverit of Oai0nl,Pblica tions In Amegrican; Arhaolg and ethnology. ~ ~ ~ ~ g U-,iersity mvof Pennsyii (ierity) ueum, Althropo- logial Publicatio, s t :United States National Euseu;- ; Reports Proceedings. 'niversity :'of ' Washinton Publicat fo i,n Anth p lojy Zeitaliit fir Bt 0tno. PERUVIAN CUMBROUS BOWLS BY ISABEL T. KELLY UNIVERSITY OF CALIFOIRNIA PUBLICATIONS IN AMERICAN ARCJHAEALOGY AND ETHNOLOGY Volume 24, No. 6, Pp. 325-341, 1 figure in text Issued April 18, 1930 UNIVERSITY OF CAL1FORNIA PRESS BER-KELEY, CALIFORNIA CAMBRIDGE UNIVERSITY PRESS LONDON, ENGLAND PERUVIAN CUMBROUS BOWLS BY ISABEL T. KELLY The problem of this paper is the unity of a certain bowl type wide- spread in ancient Peru. The series studied includes 141 bowls of relatively simple shape and decoration. Because of their unimpressive character, few have found their way into museum displays, and still fewer into the literature. The bulk of the material studied is in the University of California Museum of Anthropology. Field Museum of Natural History has very kindly loaned some 15 specimens, and the American Museum of Natural History has obligingly supplied several sketches. The bowl type which this paper seeks to establish is, for the sake of convenience, referred to as Cumbrous. This term is, in a general way, descriptive of the ware as well as the design, for both tend to be somewhat heavy and ponderous. Our concern will be, for the most part, with decorative features, but a word may be said of the ware and of the shape of typical speci- mens. The vessel is usually a gently rounded, open bowl or dish with a direct rim (i.e., not outcurved or incurved). The majority may be described as hemispherical, but some, especially those from Nazea, tend rather toward the conical in cross-section. The bottom is occasionally flattened, and one sometimes gets a definite base. This is not a southern (Nazea and Ica) trait, but is noticeable in bowls from Chancay and Ancon.1 Bowls from Moche seem always to have a tripod2 or a pedestal3 base; those from Supe4 often have the latter structure. The ware itself may best be described by terms such as heavy or dull. Compared with products of the classic periods, such as Proto- Chimu or Early Nazea, it is definitely coarse and thick. It is, however, 1 As in Strong, W. D., The Uhle Pottery Collections from Ancon, Univ. Calif. Publ., AAE, 21: pl. 43e, m, p, 1925. 2 Kroeber, Arch. Explor. in Peru, Anth. Memoirs, Field Mus., 2: no. 1, pl. 5, fig. 5, 1926. 8 Hrdli6ka, Some Results of Recent Anth. Explor. in Peru, Smith. A?isc. Coll., 56: no. 16, pl. 1, 1912. 4 Kroeber, The Uhle Pottery Coll. from Supe, AAE, 21: pl. 731, m, n, 1925. 326 University of California Publications in Am. Arch. and Ethn. [Vol. 24 well smoothed, especially on the interior, but the exterior sometimes shows traces left by the smoothing implement. In spite of smoothing, these Cumbrous bowls are characteristically dull and unpolished, features which throw them into sharp contrast with the invariably polished products of the classical periods. As practically all the specimens are whole, it has not been feasible to examine the paste in cross-section. However, the texture seems smooth, seldom granular on the surface as one might get were excessively coarse tempering employed. These bowls may be slipped or unslipped, usually the former. The unslipped range in color from terra cotta to light dusky orange, sometimes with dark firing clouds. A heavy white slip is distinctive of some central coast bowls, prin- cipally those of Chancay. This is undoubtedly associated with the Black-on-White style which at one period was prevalent in that region. The foregoing evidence in itself would never serve to set these bowls apart as a type, but, taken in conjunction with a certain con- stancy of decorative features, the evidence is fairly impressive. The remainder of the paper will concern itself with an analysis of design, seeking by this method, to establish the unity of the series. The interior of the Cumbrous bowl is always decorated; the exterior is always undecorated, at least unpainted.5 The interior decoration takes many forms, but it usually consists of stop" designs which cut the rim of the bowl and which are, at the same time, pendant from the rim. Quartering and halving of the field prevail wherever the ware is found, but one occasionally meets with three or five-zoned decoration. The principal design elements are geometric- straight lines, arcs, stepped triangles, dots, zigzags, etc. These designs are frequently painted in several colors, usually three, rarely more, and are, more often than not, outlined in black. The designs are symmetrical, with minor exceptions. Figure 1 gives a schematic representation of the entire series. Sketches and loan specimens, as well as published data, have been included. In other words, the series is complete. The general loca- tion and period are indicated on the margin of the plate. The arrange- ment is geographical and chronological. The sites from north to south are entered downward, and where periods are known, the earlier specimens are entered first in their rows. A numerical listing 5 Some few bowls, from Supe, for example, have exterior decorative impressions, Kroeber, AAE, 21: pl. 75j. 6 Design elements which are discrete or spaced borderwise are known in art as stop designs. Day, Lewis F., Pattern Design, 215, London, 1923. 1930] Kelly: Peruvian Cumbrous Bowls 327 at the end of the paper gives the museum number, provenience, and period. When the period is doubtful for some reason or other, the bowl is placed with the group to which it seems to adhere artistically and its doubtful character indicated in the list. The chronology fol- lowed is that established in previous papers of this series.7 TABLE 1 FRRQuFzQU or DrsIGN MOTIFIS Q C ) 3 C . C ) C)q 7 18 21 16 1 1 2 4 1 30 4 32 2 1 1 Quartering ..............................213 11 7 .... 1 1 1 117 2 26 1 1 .... Halving .53991.... 5 1.... 7... 1.1 .... .... .... Three-zoned decoration. ................. ....1 . .... .... .... .... .... 1.... 1 .... .... .... Five-zoned decoration.1.1 .... .... 1 .... .... Center circle with radiations ...................... 2 2 2 1 .... .... ... 1.... 3 1 .... Band around rim (decorated) .............. 4 .... 2.1.. 4 2 .... 3 2 7 .... .... Band around rim (undecorated) . .................5....... .... .... .... .... .... ....5.... 3 .... .... Lines (except block radiations) .................. 2 9 6 6 1.11 .... ....9. Lines compose unit ............. .... 7 5 1 .... .... 7.... 5 .... .... Arcs (including variants) . .................. 3 2 10 9 .... 1 1 3 .... 15 2 22 2 .... Stepped triangle ........... 2 2 59 .... .... 1 1 1 10..16 .... .... Steps within arcs ...........6 .... .... ....16 ....1 .... 4 . 15 .... .... Subsidiary arcs. .......... .... .... ....2 .... .... .... .... 4 2.... .... .... Dots .. 7 8 73.... 1 1 1 .... 8 3 15 2 .... Circles .............. 2 7 3.... ....1 .... .... .... .... 6 .... .... Crescent, dot within .............. 3 1 .... 1 .... .... 1 .... 4.... .... .... Zigzags and wavy lines ................. .... 3 2 4 . ... .... .... .... 1 4 8 .... .... .... Design outlined in black .................. 5 5 11 9 1 1 1 .... 10 4 26 2 .... Black design on solid color .................... 2 .10 5. 3 115.... 4.... 1 .... Conventionalized representation ................ 3 1 4 ... 1 . 6 .... 5 .... 1.... Human face ........................ 2 1 3 ...1 .... .... .... 2 1 .... Cat .1.... .... .... .... .... .... .... .... 1 .... .... Bird .. .... .... 2.... .... .... .... .... .... 2.... .... .... Fish . .... .... .... .... .... .... .... .... ....3 1 .... .... Llama .. .... .... .... .... .... .... .... .... .... .... .... .... ....1 At first sight, the various designs may not seem closely connected, but it is possible to arrange them in overlapping series. It is possible, furthermore, to demonstrate their genetic relationship by analysis into constituent elements. Table 1 lists the traits more or less common 7AAE, 21:1-332, 97 pls., 1924-1927, and 24:1-46, 1927. 328 University of Caliform&ia Pubicatiot in Am. Aroh. and Ethn. [Vol. 24 to the whole area. The first items deal with the plotting off of the design field. Quartering is the most common, occurring in 60 per cent of all cases. Halving is next frequent, with something like 25 per cent. An uneven number of segments is rare, but appears from Supe, Cafnete, lea, and Arequipa. The interior of the bowl is sometimes quartered or halved by a center circle with radiations to the rim. This feature is not strong numerically, but is widely distributed. The same holds for a border or band inside the rim. It may or may not bear decoration. Lines, aside from those radiating from a center circle, are a favorite motif and usually extend from the rim toward the center of the bowl. A number of straight lines may compose the design unit, as in several specimens from Supe and Ica, or they may be combined with other elements such as arcs or stepped triangles (figs. 26, 36). The motif which, for the sake of simplicity, will be designated as an arc, is in geometric terms, a lunette. It is the area between the rim and an intersecting arc or pendant loop and is one of the out- standing features of the Cumbrous design complex. The arc may be solid or may contain concentric arcs, stepped triangles, or lines. Cer- tain variations of the arc occur; the outline may be skewed (fig. 72); it may be angularized (fig. 83); it may be flattened (fig. 37); the inclusion of the latter cases cannot be challenged, as they pass by transitions into the normal arcs. Where a wide band crosses the face of the bowl, the residual areas on either side, of course, tend to appear arc-shaped owing to the contour of the vessel.8 For the most part, however, these automatically produced arcs are undecorated and apparently have no entity of their own. For this reason, they are not counted in the tabulation of arcs. When one gets conven- tionalized human representation, the arc often serves as the facial outline (figs. 6, 63). The stepped triangle, either isosceles or right angle, is another standard motif. It is usually solid black and has from three to four steps on a side. The plain-sided triangle is rare. This is of some interest, as the painting technique would not necessitate the jogged character. The three motifs mentioned so far, the line, the arc, and the stepped triangle, may be characterized as the fundamental design 8 Kroeber, AAE, 21: pl. 73m, 1925. Kelly: Peruvian Cu'mbrous Bowl3 elements of the complex. One hundred six, or 75 per cent of the 141 bowls, show one or more of these features. In addition, at least one of these three elements is represented from every site with the excep- tion of Chichillani and Humahuaca, two southern highland sites, each with a single specimen. That from Chichillani shows a faint sugges- tion of arc treatment, but its inclusion is justified because it exhibits the typical quartering. The Humahuaca bowl (fig. 138) departs even more markedly from the norm, but it is so similar to an Arequipa specimen (fig. 135) that its relationship to the series cannot be questioned. The evidence for connection between the various regions is further strengthened when more specific traits such as steps within arcs and subsidiary arcs (fig. 49) are shown to be held in common. Numer- ically, these are unimpressive, but their mere presence is suggestive. In addition, dots, circles, zigzags, and wavy lines are generally distributed. As before suggested, the design may be painted in one, two, or three colors-black, red, and white- in addition to the ground color. Seventy-five bowls, or over 50 per cent of the total number, have the design outlined in black. Excepting the bowls painted in black only, one finds that 75 per cent are outlined in black. Excepting the sites which yield only black monochrome designs and hence no outlining (Chincha, Cafiete, Chichillani), one finds black delineation from all regions but Humahuaca. Here there is a suggestion of outlining, but the sketch figured by von Rosen9 does not allow of definite interpretation. There is a decided preference for the geometric, and conven- tionalized representation occurs in a scant 19 cases, scattered from northern Peru southward into Bolivia. The tabulation seems to sug- gest that conventionalized human faces center in the north and animal figures in the south. It remains for further data, however, to confirm or refute this suggestion. The above material may be summarized as follows: 1. There is a constant adherence to the device of quartering or halving the design field. 2. There is an impressive recurrence of three motifs, the line, the arc, and the stepped triangle. 9 von Rosen, B., Popular Account of Archaeological Research, fig. 237, Stockholm, 1924. 1930] 329 330 University of California Publications in Am. Arch. and Ethn. [Vol. 24 3. Various other design elements such as the step within an arc, the subsidiary arc, the dot, the circle, etc., occur widely. 4. Finally, it is possible to arrange the designs in continuous, over- lapping series. The foregoing discussion demonstrates, I believe, the unity of the series and reveals common elements of bowl design for the entire Peruvian coast, giving, moreover, definite suggestion of highland adherence. These resemblances cannot be explained on the principle of limited possibilities, because the possibilities of interior bowl decora- TABLE 2 DESIGNS CHAR.AOTERISC OF SOUTH AND NowT 7 18 21 16 1 1 2 4 1 30 4 32 2 1 Subsidiary s t e p s ............ ...................... .... .... .... .... .... .... .... .... .... .... .... 13 .. .... .... Quartering by arcs (4 arcs, 4 pairs, or 4 and subsidiary arcs) ........................ .... .... .... 2 .... .... ... 1 ... 9 2 18 .... .... .. D iam etric band .............................. .... 6 6 .... ...1 .... .... .... .... .... .... .... .... .. T w o arcs .......................... ...... 2 2 9 7 .... , 1 1 1 .... .... .... ... 1 .... .. Steps, straight lines sam e unit ..................... .... 2 5 1 .... .... .... .... .... .... ... 1 .... .... .. Scroll (not interlocking or conventional-15 ize ............... .... .... .... .... .... .... .... .... .... .... .... .... .... In terlocking scroll ............ .... .... 5 .... .... 1 .... .... .... .... .... .... .... Sol id arcs ................... .... .... 5 1 .... .... 3 .... .... .... .... .... .... WS ite des ign on soli d color ................... .... 11 .... .... .... .... .... .... 1 .... .... .... .... . tion are many. Various combinations of motifs, colors, and arrange- ment might occur. The arrangement alone might be all-over, centered, or bordered, with many variations of each. Cumbrous bowl decora- tion is not at all obvious, or it would certainly appear in other regions of the world where interior bowl decoration has been developed. In the Southwest region of the United States, for example, bowls are plentiful, but the decoration differs fundamentally from that of the Cumbirous type.10 There is no resemblance in color, arrangement, or 10 A striking exception to this is found in Kidder, A. V., Pottery of the Pajarito Plateau, Mem. Am. Anth. Assoc., 2: pl. 26, figs. 3, 6, 1915, in which one gets a close parallel to the Chancay interlocking scroll. Figure 8 of the same plate is, moreover, suggestive of the Cumbrous arc. These isolated cases are the more noticeable as they vary so markedly from the norm of South- western material. 1Kefly: Peruvian Cumbrous Bowls final effect. And this is the more marked as Peru and the Southwest share simple motifs such as the stepped triangle. Cumbrous bowl decoration, as a type, is the more distinct when brought into com- parison with bowl decoration from other areas. Table 2 lists traits which show a tendency to group either in the north or south. Subsidiary steps, or steps appended either side of an arc ( figs. 107, 130), are confined to Nazea. Likewise, the arc as a quartering agency is noticeably stronger in the southern sites. On the other hand, certain traits tend to cluster in the north. They are-the diametric band (figs. 19, 39); preference for two arcs instead of four; the combination of steps and straight lines (fig. 13); the simple scroll (fig. 22); the interlocking scroll (figs. 28, 29) ; solid arcs; and white monochrome painting. One can draw distinctions still finer and characterize each well represented site. When more than one period is represented, there is usually a corresponding stylistic variation. Chimu area.-A few specimens of northern, but locally doubtful provenance, are grouped under this heading. Undoubtedly, they con- stitute the least uniform and the least typical lot. In general, these specimens are of the Middle period; there are no Proto-Chimu speci- mens which conform. The following are included: two Middle period bowls from Chicama which are figured by Hrdlicka;"- two Peabody Museum specimens, presumably Middle in period, from either Viru or Chicama ;12 one bowl, period uncertain, purchased at Trujillo and figured by Kroeber ;13 two black-on-white bowls from Site A, Moche, which are definitely Middle in period. As before mentioned, these are far from typical, but certainly bear some relationship to the series. All are halved or quartered; some show the center circle with radia- tions; others show arc variants and conventionalization. Supe.-Here, all specimens are from the Middle period. Supe has no true arcs and specializes in scrolls and monochrome white on red or terra cotta. The simple line unit (fig. 7) and the diametric band are also frequent. The latter often contains a running scroll pattern (figs. 20-24). Chancay.-Two Middle styles (Three-Color-Geometric and Epi- gonal) and a Late style (Black-on-White) yield typical specimens. II Op. cit., pI. 1. 12 Kroeber, Moche, AAE, 21: pl. 69a, by 1925. 13 Op. Cit., Field Mus., pI. 11. 1930] 331 332 University of Californma Publwations in Am. Arch. and Ethn. [Vol. 24 The former are represented by 12 bowls and the latter by 9. The Middle and Late styles overlap, being associated in graves of one cemetery; but the Late style is also found pure in other cemeteries. There is a definite stylistic distinction between the two Chancay periods. The Middle period makes use of the step-line unit, the inter- locking scroll, the center circle with radiations, and the arc as a human face outline. In the Black-on-White style, attention shifts to the solid arc variant and the center band. The step and interlocking scroll persist, but feebly. The tendency toward solid black designs is quite in keeping with the general Black-on-White style and probably is a concomitant of it. Ancon.-Thirteen typical bowls are Middle Ancon and three Late Ancon. The periods here are not set off quite as sharply in style as at Chancay. The Middle period exhibits the Cumbrous decoration complex in its most generalized form. There are arc, step, and line units and a tendency for steps to occur within arcs. The ensemble impresses as being quite close to Epigonal Ica. The later period, although scantily represented, exhibits no marked differences. One late bowl (fig. 59) uses the same motif as is found on a Chancay Three-Color-Geometric jar. Another late specimen has four solid arcs. In view of the general similarity of Ancon and Chancay pottery, one would expect much closer agreement than is actually the case. There is not one instance of the typical Chancay interlocking scroll and only one case of a solid arc. For the Late period, this might be attributed to the paucity of numbers, but the Middle period is quite as well represented numerically as is that of Chancay. Soutth central coast sites.-Material from this region is noticeably meager. The few specimens show the usual straight line, arc, and step motifs. One of the two bowls from Pachacamac is after Uhle.1' It is of the interlocking scroll type and, from the picture at least, is indistinguishable from Chancay examples. The other Pachacamac bowl is from the American Museum of Natural History (B-472) and shows the typical arc, step, and step within arc elements. Ca-nete is a central coast valley whose chronology is not decisively placed, but the material drawn upon is relatively early. The four sketches shown are of bowls in Field Museum. While they are not 14 Pachaeamae, pl. 7, fig. 6, 1903. Kelly: Peruvivan Cumbrous Bowls thoroughly typical, they do employ the arc and stepped triangle. Several other Early Caniete bowls, not figured here, suggest the complex, and might conceivably be regarded as variants. Ica.-Ica is represented by 30 bowls from the Nazcoid, Epigonal, Middle, Late, and Inca periods. None of the Nazea style from Ica can fairly be included in the type. Nazcoid (1 case) and Epigonal (7 cases) yield the most typical specimens. All have arcs and several have, in addition, step and line elements. There is a noticeable similarity between the Epigonal Ica bowls and those of Middle Ancon.15 One finds stylistic divergence beginning with Middle Ica. Artistic- ally, there is deterioration. Step designs continue. The are also continues, but in modified form. It becomes angular-like an isosceles triangle with the base on the rim of the dish. This arc variant is sometimes filled with lines parallel to the sides and crossing each other at right angles (figs. 82, 83). This cross-hatching feature is fore- shadowed by an Epigonal design (fig. 72). A yet more radical variation of the arc occurs in Middle Ica. The outline is omitted and a triangular area built up of rows of super- imposed dots or diamonds. Lines are more frequent than in the Epigonal. They occur, 5 or 6 to the unit, and run in a short distance from the rim. Several bowls show no specific affiliation with the earlier style aside from halving and quartering. This is accompanied by several new motifs-conventionalized cat figures, small rectangles, and L-shaped units. All this seems characteristic of the Middle and Late Ica geometric or textile-like style. The five representatives of the Late and Inca periods show quarter- ing by lines. One has the familiar center circle with radial lines. All have conventionalized animal figures-three of fish and one of a bird. One bowl from this late era has an unoutlined arc variant built up of small diamonds. As a whole, these Middle and Late Ica bowls are manifestly far afield from the typical Epigonal forms. Yauca.-Yauca is a site south of Ica from which we have four bowls. Two are quartered by arcs and have, moreover, subsidiary arcs. The other two (figs. 100,102) are not particularly characteristic. All of the Yauca specimens have wavy lines. Nazca.-Bowls showing typical decoration are abundant in the Late Ica period from Nazea. Definite evidence of the complex appears 15 Ancon, for example, is the only northern site which has subsidiary arcs and steps within arcs. 1930p] 333 334 Uniiversity of Ca2iforma Publictiotns in Am. Arch. and Ethn. [Vol. 24 earlier in certain Nazea Y-Epigonal pieces, and various other speci- mens of the same period are strongly suggestive of the complex. A Nazea Y bowl shown in plate 14f of Gayton and Kroeber16 has a step and line design on the outside. Plate 14e of the same paper shows a rim segmented by lines and skewed arcs. Two Field Museum bowls (170483, 170485) which are classed as Nazea B, verging on Y, are quartered by stepped triangles. However, as they are four-color, and as the outside rim is circled by a fret design, they are not included in the series. Quartering appears as early as Nazea A. One gets two wide lines crossing at right angles and dividing the bowl into four sectors. The triangular sectors thus formed are sometimes filled with human faces. A Nazea A bowl of this type from Ica is figured in plate 28o of 'Kroeber and Strong.17 The general character of these early Nazca bowls, however, does not conform to the type under consideration, and they have not been included in the series. The Late Ica group from Nazea has a considerable number of bowls which cover almost the entire range of the complex. One gets quartering, but, surprisingly enough, no halving. The center circle with radiations is present, but rare. Arcs, steps, and lines are more than plentiful. The latter somnetimes run in from the peak of the arc instead of cutting the rim. About half the specimens have steps within arcs. In addition, there are several special features which seem particu- larly associated with this Late Ica period at Nazea. Wavy lines are abundant. Subsidiary steps, or steps attached on either side of an arc, are confined to Late Ica at Nazea. Another device which is prac- tically limited to this area and age is the filling of all free space with series of dots, circles, lines, etc. (figs. 125, 126). Highland.-The University of California Museum has little high- land material and provides only two Arequipa bowls for the series. One of these has the rim cut by five arcs, and the other exhibits an aberrant form of the center circle with radiations (fig. 135). Of the other highland specimens, one is from Chichillani, Bolivia,18 and was obtained from a sketch furnished by the American Museum of Natural History. It is quartered by the figures of four llamas and 16 AAE, 24, 1927. 17 AAE, 21, 1924. 18 If this is to be identified, as seems likely, with the Chinchillani of Bolivian maps, it is near the Chilean border, slightly south of a line due east of Arica. Kellfy: Peruvian Cumbrous Bowls by four rounded " V " elements which segment the rim. These V-shaped objects are faintly reminiscent of arcs, and the bowl unques- tionably falls within our series. The remaining dish (fig. 138), a "ground" find from Quebrada de Humahuaca, near Juyjuy, Argentina, is figured by von Rosen.19 It is in no way typical, but bears a strong resemblance to one of the Arequipa specimens (compare figs. 135, 138). Material from the inland districts is ra.re at best, but in time more will undoubtedly come to light. It would be exceedingly interesting to be able to trace an unbroken distribution of bowl decoration from coastal Peru into the Andean region. This completes the series. It should be apparent that, despite the relative uniformity, an appreciable amount of local variation can be detected. SUMMARY The foregoing data may be summarized as follows: 1. A bowl decoration complex was shared by all of coastal Peru and probably by the highland region as well. 2. When the complex continues through. several periods at a site, the basic elements rema.in uniform, but certain stylistic changes are apparent. 3. We have few bowls from the Chimu area, none of which are thoroughly typical and none of which a-re early. 4. Material from the central coast is scant, but there is a close resemblance between Middle Ancon and Epigonal Ica, despite the intervening area. As regards historical development, the situation is far from clear. Dr. A. L. Kroeber has very kindly supplied the chronological basis for table 3. To this table have been added the specimens attributed to each period. Numbers standing free represent specimens which can be assigned definitely; those in parentheses indicate specimens which proba.bly belong in the period. The chronology of Supe, Chancay, Ancon, Ica, and Nazea is reasonably certain, but the other sites have not been so thoroughly equated, and this uncertainty is indicated by the dotted lines. Assuming, however, that the chart is approximately correct, it becomes apparent that there is a definite bunching on, and imme- dia.tely following, the Tiahuanaco horizon. With the exception of Chancay and Nazea, the complex tends to dwindle out in the later 19 Op. cit., fig. 237. 1930] 335 336 University of CaJifornia Publicatio in Ain. Arch. and Ethn. [Vol. 24 periods. At Chancay, the style may have been given impetus and new character by the Black-on-White technique, although, in that case, one might normally expect the same to hold for Ancon, which shared this style. The specimens from Late Ancon are, however, too few in number to make the absence of a trait significant. It is impossible with the few data in hand, to essay anything in the way of historical reconstruction. Such meager evidence as there is might seem to hint at a southern origin. We have precedence for quartering in the early periods of Nazea and Ica; the bowls from this region share the general basic elements but have, in addition, strongly specialized traits such as wavy lines, subsidiaiy steps, etc. Yet the significance of the early appearance here is partly nullified by the fact that we have no bowls at all from early northern and central coast sites. In fact, aside from Proto-Chimu,20 Nazea is the only early style yet discovered. As historical reconstruction is not feasible, all that can fairly be said is that the complex which is characteristic of Cumbrous bowls is Epigonal or Tiahuanocoid in feeling and in period of first appearance. It is, perhaps, one heretofore neglected element of the general Tiahua- nocoid complex. Further data, especially from the highland districts, will do much to clarify the situation. The fact that an intrusive type of ware could impress itself on a series of distinct local styles and persist practically unchanged in essentials for what must have been several centuries, is a matter of some interest. Something in the dynamics of Peruvian coastal cul- ture enabled it to assimilate successive styles such as those of Tiahua- naco and Inca and to recast them, but in such a way that the intrusive type persisted in its essentiaUly distinctive features. 20 Which is probably later than Early Nazea. Kroeber, Field Museum, op. oit., 20. Explanation of Table 3 Numbers standing free represent bowls which can be assigned definitely to a period. Numbers in parentheses indicate specimens which probably belong within the period but because of incomplete data, etc., cannot be accurately placed. So little is known of the chronology of Yauca and the highland region that these sites have been entered without any attempt to equate them chrono- logically. As a consequence, the number of specimens is entered below the site name rather than on the chart. Five Nazea bowls were designated by the collectors (Kroeber and Schenck) as "Epigonal or Late Ica." On the time chart, these have been grouped arbitrarily with the Late Ica bowls as this seems the period into which the characteristic Cumbrous bowls fall at Nazea. 1930] Keily: Perwiian Cumbrous Bowls TABLE 3 OOoUaRRCE BY PmrOD Beginning End of of Tiahuanaco Tiahuanaco Inca 15 influences influences influenoes A. Tiahuanacoid 2+(5) Middle 18 Epigonal and 3-Color Late Chancay - 6+(6) - 7+(2) Middle Ancon Late Ancon 12+(1) 3 Proto-Lima 1 Cajamarquilla - -11 - - Late (2) I Early Ca iete 4 Late dhincha Nazcoid Epigonal Middle Late Ica (Inca) 1 7 17 2+(2) I 1 Nazca Y-Epigonal Late Ica I (Inca) 2+(2) 7+(21) For explanation see opposite page. 337 630 .D. Chimu Area Supe Chancay Ancon Nieveria Cajamarquilla Pachacamac Caniete Chincha Ica Yauca 4 Nazca Arequipa 2 Chichillani, Bol. 1 Humahuaca, Arg. 1 338 University of Califomaica Publications in Am. Aroh. and Ethn. [Vol. 24 EXPLANATION OF FIGURE 1 Unless otherwise indicated, the numbers below refer to specimens in the University of California collection. AMNH stands for the American Museum of Natural History; FM, for Field Museum. When material has been illus- trated, reference is cited. All references are to plates in papers of this seri" (AAE) unless the author's name is given. When known, site and period are indicated. Spec. No. Published 21:66h Kroeber, 1926, p1. 5, fig. 5; pl. 11, fig. 4 21:69b 21:69a Hrdlicka, 1912, pl. 1 MOCHE, Site A MOCHE, Site A Bought at TRUJILLO VIRU or CHICAMA VIRU or CH1CAMA CHICAMA Middle Period Middle Period Probably Middle Probably Middle Probably Middle SIUPE, San Nicolas Middle Supe CHANCAY Site is not known for figs. 30, 31, 35. Remainder are Site C. Figs. 25, 27, 29, 33, 34, 36 are 3-Color or Epigonal. Figs. 26, 28, 30, 31, 32, 35 are probably the same. Fig. la lb 2 3 4 5t 6f 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 4-2550 4-2551 4-7056 4-7163 4-7373 4-7367 4-7867 4-7374 4-7356 4-7375 4-7067 4-7379 4-7059a 4-7211 4-7310 4-7126 4-7121 4-7366 4-7364 4-7363 4-6638 4-6544 4-6509 4-6543 4-6583 16-917 16-1161 4-6606 4-6605 4-6618 4-7052 4-6577 21 :78d 21:75- 21 :73n 21 :73o 21:73k 21:78k 21 :78f 21:731 21 :73m 21 :84c 21 :84a 21:83a 21:83c 21:83c Kelly: Peruvian Cumbrous Bowls CHANCAY-Continued Published Site A 21:82f Site C Site B La Ninia La Nifia Site B 21:82g Site C Site C Site A 21:47i 21:44r 21 :43m 21:43e 21:43p Figs. 37, 38, 39, 42, 43, 45 are Late Chancay. Figs. 40, 41, 44 are probably the same. ANCON Site P except fig. 50 for which site is not known. Site T Site B Site A Site P } Figs. 46-49 and 51-58 are Middle Ancon. Fig. 50 is probably the same. Late Ancon 62 4-9197 21:96a 63 AMNH 3417 64 AMNH B-472 65 * Uhle, 1903, pl. 7, fig. 6 66 4-3751h 21:fig. 7f 67 68 69 70 FM 169618 FM 169619 FM 169742 FM 170252 I NIEVERfA21 Proto-Lima (Str. A) CAJAMARQUILLA PACHACAMAC PACHACAMAC "Pre-Inca" (Late?) CHINCHA, Site C Late Chincha CARETE, Cerro del Oro Early Caniete 21 A Cumbrous bowl from Nieveria is figured in Uhle, M., tber die Friih- kulturen in der Umgebung von Lima, Int. Cong. Am., 1908, 16: fig. 19, 1910. This bowl has quartering, wavy lines, arcs, steps, and steps within arcs. Unfortunately this reference was not encountered in time to permit its inclu- sion in the figure and tables here presented. It is thoroughly typical of the Cumbrous complex and serves well a-s a link between the northern and southern coastal sites. Fig. 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 Spec. No. 4-6444 4-6524 4-6490f 16-1016 16-1015 4-6480a 4-6694 4-6649 4-6431 4-6151 4-5920 4-6100 4-5996 16-916 4-6049 4-6117 4-6048 4-6021 4-6180 4-5919 4-6047 58 4-5730 59 60 61 4-5598 4-5589 4-5908 1930] 339 340 Uni-versity of Caiformia Pubicati&ns in Am. Arol&. and Ethn. [Vol. 24 ICA Spec. No. 4-4495 Published 21:29e and Uhle ,= 1913, pl. 11, fig. 3 72 4-4470 21:30g and Uhle, ibid., fig. 5, fig. 8 73 4-4545 21:301 4-4486 4-4552 4-4478 4-4538 4-4547 4-4604a 4-4603a 4-4603 4-4603b 4-4602c 4-4602a 4-4602b 4-4318o 4-4580d 4-4580c 4-4318c 4-4580e 4-4580g 4-4580h 4-4581 4-4580a 4-4613 21:30m 21:fig. 11 21:30h Ocucaje 21:31c Chulpaca Ocucaje Ocucaje Chulpaca 21:35b Ocucaje 95 4-4560 4-5304 16-874 4-5236 4-5392 Site T Site T Site T A Nazcoid (Nazca Y at Nazca) E E E Epigonal E E E E C C C C C M Middle Ica C C M C C C C C C G Probably Late Ica Late Ica Probably Late Ica Late Ica Inca 22 Uhle, M., Zur Chronologie der alten Culturen von Iea, Jour. Soc. des. Am. de Paris, n.s. 10, 1913. Fig. 71 74 75 76 77a 77b 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 96 97 98 99 Kelly: Peruvian Cumbrrous Bowls Spec. No. Published 4-8183 4-8180 4-8181 4-8182 FM 170618 FM 170155 FM 170459 FM 170583 4-8763 4-8765 4-8764 4-8540 ,FM 170127 FM 170128 AMNH 41.0-886 FM 170433 AMNH 41.0-885 FM 170612 FM 170129 FM 170606 4-8766 .4-8761 FM 170437 4-8772 4-9005 4-8768 4-9026 4-8767 FM 170620 FM 170809 FM 170939 FM 170941 FM 171002 FM 170938 FM 170808 AMNH 41.0-768 YAUCA 24:18f 24:18e 24:18d k NA2 Epigonal Probably Nazca Y-Epigonal Probably Nazca Y-Epigonal Epigonal Probably Late Ica Late Ica Late Ica Late Ica Epigonal or Late Ica Epigonal or Late Ica Probably Late Ica Late Ica Probably Late Ica Epigonal or Late Ica Probably Epigonal or Late Ica Epigonal or Late Ica ZCA Late Ica Probably Late Ica Late Ica Late Ica Probably Late Ica 4-8062 Bought at AREQUIPA 4-8063 Bought at AREQUIPA AMNH B-7497 CHICHILLANI, B6LIVIA von Rosen, QUEBRADA DE HUMAHUACA, 1924, fig. 237 ARGENTINA Fig. 100 lOla lOlb 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 1930] 341 AM. ARCH. & ETHN. VOL. 24. NO. 6 [KELLY: FIGS. I.138J 0? 3 la-6 North Coast (generally Tiahuanacoid) 5 6 9 10 11 12 13 14 15 to.. 16 . @. 0 0 0 *0 17 7-24 Supe (Middle Period) 0 ? 0 , 21 22 23 24 27 0 37 38 31 WW]DO? 39 40 41 42 43 33 w V 44 0 . . 00 45 35 25-45 Chancay 25-36 Three Color and Epigonal 37-45 Black-on-White c3 47 48?3 0 56 O?II?E3 49 50 51 57 58 59 ,52 0 60 53 54 ?II3 55 46-81 Ancon 45-68 Middle Ancon 69-61 Late Ancon A?4 61 0 71 62 0 72 010 . 0 0 73 ?QCJ(J3. 68 159 7n 740 00 75 76 77aC ' 77b0 780 A7 ? 79 800 62-70 Central Coast 62 Nieverfa (Proto-Lima, Strain A) 63 Cajamarquilla 64-66 Pachacamac 68 Late Chincha 67-70 Early Caflete 71-99 Ica 71 "Nazcoid" (=Nazca Y) 72-7Th Epigonal 78-94 Middle Ica 95-OS Late Ica 99 Inca 81 82 83 84 85 86 87 88 89 91 92 93 94 95 4Q ?b ,? ?p * 4 0* 96 97 98 99 100-102 Yauca 100 lola 105 106 107 117 118 127 128 135 136 108 109 0. 119 120 129 130 ED 137 138 110 111 113 103-134 Nasca 103-106 Nazca Y-Epigonal 107-134 Late lea 121 131 132 133 134 135-438 Southern Highland 135-136 Arequipa 137 Chichillsni, Bolivia 138 Humahuaca, Argentina 7 'a I ? ( 8 2 0 20 36 46 90 .1? gO 103 1 01 b 104 102 114 115 116 125 126 II . I~i,, I1.I~ .-U! f . lV,~ ..$ITYOFI -J~ 1 OA.TBN4 tY ' LIIOI4"~tIO SII ..-~~~DP Wr E T~ A.N ,, 1- W~ 7~OP0,LOGTI~~ I- --1 . I I i pbIcations ~ ,,II delig it. rhaolgia as I etII.oogcl ujet isudune t, diecio o te eprten oI Atlr'o1 .yaI su .-i3 . l~agecrth p _bicaiou o beadrse to the. UNIVBUITT O CALII1LIA EES$ - Ed-,Itr. rIce - Volume, 11,~, ",, $4 2_; V'o1ume_ 2 to 11, ic-siro, $3 .50 each; I frou .1 voumsI~ -.5 &0 e.; olm -, # I4.50, mipIeIent"I25 cens;voume27 #1 l.- . 1 w -- Voum s 4 25,. ., ant_ 28 In prorOss- , g1nnI g *tt'' v'oliI Vo.1~1 yh ft~Suhr IraMw~ y8 .Brt.,p .8. Marol-I,. - ...wIth3Ip.INovembe, 1919..... -I-..1 I~:- 5. Yuo1 GegI -h ,by'I.~ T~.. aema. p. 1-44,el .e --lW I, tor -,' 8. I he CtahIf Inila , b-'L -le o- er Pp 1 0.O_ Apdl.', U1920. .~. .7 7.Ihe utoiogaph ~fa -m1-ag-IhI,n lIPal Edin - p. 8147$ 8.~ ulanTrbe ofth Lwe Qoor, by -- Z Kroebr P. 4 I5I453. I" _gu t 1 2 .....,....'i 2 In4lex, pp.:487491". N~ 1 Pp.- I 7.25, 3ig r I n tox- Peb -za, I192." f.25 - - Pp 373 I68, pate 20 i82 1ukee I `-W.v.e4 e, 1_-"I"".925 ,I.4- - index, p. 409-410 Vo. 8 1 C1fo4 MOKishp:e8.ooge, y darIW IoGiod.Pp128,.~ N " ~ih 2 u~.Z)cebeI 129I.. .....I4) Index, pp. 407-409. -, - -. - ~~~~~~~~~~~~-.1 1 ~-_ .-. .-'I Ilc2flue i et.-Dcebr 12 -&-OG, 2. Explo lortinsat- 4ch, y a Uhid . P.. 6j9,1k gur l text. Ho. a4d2 n n.coe ., SeptImber-l1924 ; ..... 1.89 - .. A - 4.~ t a - _. ~ I UNIVEHBSITY -OP. OA3IJOZA L 5.- The ec,by L ber. - -.4 plates 50 -8-9,, 5 figures -n text. 6 B. he Uhlb; Pottea 0olecs froni. up lA.L KSb P- - 'PN b ad-6 i one cotvt. Deeme, 9 w ............... it- 1.25 L X~~~~7 Th Uhi Potr .o;etin :frz; 7.ncy l,y A. Z.adee n Krobar. Pp. ;265--;> ; 304, plates 309&, * 6 * :: - i, test,i Z, . .60 Wo8. Te itTllt Pottery Colects .from iek. 2 r t . -. Xr4e, ~p 3713SG ;, AH&np $ j*P t tb 1 G^w ''dst- : ' vL-22. : Wlyot bra~inar ad:: be, by Gladys A. 3lclatd P6'^ >,-2 plate L :.::-. 2.i . - alffornln Anthrop~metri', bys s d Wslow; G.iffQrL Pp > 1b;; 90, plat.2-3,t - - M j I 2.25 ;~ ~~3 WXsz Tet by Grae Danber P!p. 391-443. Pe ar,1Z f .6 t 'Gr ~ ~ ~ : b" .:Evl_ >A-; :, -Gi \ -id ; W.bet s Pp 2 pate - I * ; :~~~~~~~4 46 Aug0X0 a ?== ?-; tV o riteoloAorginal Goups ion thCaiifobi ela o, yW gbt Sc;etoidk Pp. IS-4 2 g n t ove 126 . .- .0 . "The: vlryvlUe 6helmota4 (Tlul epo, y W. lgbert aSheck Pp. - <- X:0 -. 0147-22W,- plates 355, B Agk in et,1mp.; 1toemfbr, 1926 -.. 1.7 .4 Aro 4,las Ditibutibou ab A. L.->> rober Pp2-268, I. a*>arp 5. AcoiS~w Gogrpby b Or"d 6.vt~n p 9-4Zpae,6-9 : :i$$tia. , '*'t--..-*. .s...;-. .45- -.;E. X: t; ;; f. Pitcla Accu In, ia yPly RatB&io Goddard, p?p. .-38.' ,Jaau e - :y. 7. ~ote5 on tewn Li, bi' ri 3.w W. Gea . 14 Logie Pp. 339-2 April, 1928. .. ...g ... .25 * :- ; -- ;;8;. Potter-aigl ?t-th4SoirUwe% -by 3. W. (ifford.; .Pp. 853-373, - ~gi : i : ; ::\iX0I Nai. Culture in -bb 1thwp ,byA 4 L Boeber. . Pp. 876-498 J.ly Pates 6c-67. .Doo,, . 19yR ; Gd .d I0de p6p. -' .L44 if . ., 2t . U6 hi Pte Co Y cStiQ* ZromY Kasca y A g. n -and A t Vol; ''- - r .-: s d = i, : X,. .: '251 plte 9.-02 2- -gie i- te>;2,pa/es sxts,1 map. September,--1929 -*<*.-...*--,-^.. ', 4. "Th f 825-41 I 4 ,'to iii ,',.. , >- ~ ~ o 25 - 80. . 1 . Loye>loe Cave, )ty', u4swell*~4fiS*v,;te'v L. Loud and M. IL Ntlr$ _r. Pp ii - f~~ ~~2 ~~tw~Z~goi ~-,b dwnMZob Pp." ,8-27 pletes'' 09-#X"0 'Z3 4. Arha eology of the NorlS tJoauVl byW lgbert ehn :~~~~~~n .le 3. ,.wjn '.-. 289413 plte .:~2 :,"emer ,;::- aia-l.',"--;'. , 1.0; _:.'-, Vo-; 25. ^Aboriginl Society Bonthrt Caliorna,k b* 8 . i - S .; 1+ Snppemen---utbo a&Zil ld Uir5 tgo; %aori Pb Noe;, mbier, 193 , O '- -r.t.-..-..,......, - 5 VoL 2 1 a , b n L. (laQ9u. P -21, 9 fres l- tt 2 .- T fpte e vin Anciet Peb,., by Li : ''a . L - --'.\ ,. -Z .-.The G.hSt Dance of- 187Q"-;>I- Soth-enra Obifrnis",' by-' A . ', Gayoz. 1929b130.......~ * , ! '; V f " ; 5 ,0; !p p . % " e ,. w 4 M 'b u , 0,'zf; X,~~~~~~~~~~~~~~ . .-- - '- 4.- - ! ' --4 ,0