“DEPARTMENT | oF ANTHROPOLOGY — Nese ‘The yoilowine: publications dealing. with archaeological and ethnological. subj “ linder the direction of the Department of Anthropology are sent in exchange for the pit .. Cations of anthropological departments, and museums, and: for.‘ journals: ‘devoted g Ge Prices, Volume 1, $4.25; Volumes 2.to 11, ine ike $8.50 4 progress, pyeigs ona. 25: Oe a Os picoc\: an 3B, eo are and 22 392 ‘pages, BL plates: 1904-1007 Pes Se $44 ‘pages, 1905" p Suits sober sala 4. 374 pages, 10 plates, 1 map. 190 “Dy. B84. pages, 25° plates. ” 1007-1910 6. -400:p pages,: 8 maps. 1908 aN fran 2, Bancroft Library, ‘by A, Te ‘Kroeber. . Pp. 1-27. Ma: 2 “The Ethnography of: the: Cahuilia Indians, by A. L, ».. plates: 4-15: daly, 1908 Pe ee ee . wes CS ree . The Religion of the Luisefio and Dieguefio. Indians of Southe 2 Aaniee “by Constance Goddard Dubois, ae 69-186, plates 16. : 54 4 The Culture of the Liutsefio Indians, Philip tedman a UCC eo aed, plate BO.” Aumnat, 1908 a 8 ea ee ae By, a om ‘Notes on Shoshonean' Dialects” of) Sete Californi : Su wa Pe ig oe ae Pp. 235-269. * ‘September; 1909. iwichla bee terse Nee eal denchactichaeapuces a 6, The ‘Religious’ Practices of the ee WA f. TT. Water ae ae , 811888, plates 21-28," if : : eae foes Index, pp. 359-36 vale, “4, enw ‘Texts, by. Edward’ oe together with D4 oS Roland By. ‘Dixon. | “Pp, 1-235, -February 1910 an) “8 The Chumash and: Costandan. Languages, . 53). “November, 1910 Pans SANE Seek ie A ENE alas ine ees ce Ss. ‘The Nei es as the ean oe Oattornis 191 6 San, wa ‘Mission Record ‘of. the ane ‘Indians, ae “mans Pi tere pistes. 15. Nasmubecabre oe -. $. Phonetic Elements of the Mohave Language, by A. i, oe “> plates 6-20,’ ee Ne ee 21-87, December, 1912 Gilese Rs Ve 6, Papago Verb Stems, by. Juan Dolson. gens 41-263. Aug . Notes on: the Chilula ree nria of urea vee orn os |. April, 3 Obtiuia "exts, by Pliny’ Barle Ep. 289-87 Index, pp., 381-885. F : ‘~ fs gus ige, bY: y s , 4b. October, ROR eas a ee = = ‘Phonetic ngvon, PP 1 of Sa hay anguage, dS Pate UHLE POTEERY COLLECTIONS PROM CHANCAY BY A. L. KROEBER UNIVERSITY OF CALIFORNIA PUBLICATIONS IN AMERI AND ETHNOLOGY . Vol. 21, No. 7, pp. 265-304, plates 80-90, 26 figs. Isued May 19, 1926 LonpoN, ENGLAND Aes THE UHLE POTTERY COLLECTIONS FROM CHANCAY BY A. L. KROEBER Wirn APPENDIX BY Max UHLE CONTENTS as Oho bs MWh ET SEG OP a SI eae pee een Ae en 266 Reig eleOnewnire Boyes Ay By Doce tins-Sesscdieececesginesoconsvinns Vor) ate ee ee 267 ae ipec-coror and Hpigonal styles: Bite Cui. c...occccccsecssscssesescsscesecsosssnssssscsesovesusenseseavae 271 White-on-red and Interlocking styles: site Boo....00ccccccccccececeeeseeeeeeeseseveveeees 275 TIES UCU ar iS 26 Sy Sg WR ce 2 een eee en 276 ON pe be ora cd RUPP 9 Rag RN TH ae ee et ee Ee 279 LOLI OM SPO DRUM EUW OLS TY LES Gb Hrasguc sescacceotsevissccede ots ac vende sSheady coostuontenebnie es ov utteaseasiescns 283 COS SESE SVR) oa 78 RA ae Rn 291 Appendix. Report on Explorations at Chancay. By Max Uhle......00000000000..... 293 Specimen mnurabers of Objects illustrated s.. (..6..5...c.cccccccoessqeesvescseedsrsesessepivavsoneeesntenenees 304 PLATES Following page 304 80-82. Black-on-white style ware from sites A, B, C, D. 83-85. Three-color Geometric and Epigonal styles from site C. 86-87. White-on-red style ware from site E. 88-90. Interlocking style ware from site E. FIGURES IN TEXT 1-4. Black-on-white style goblets and low bow]s.........0...00ccccccccceccceeeeseseseseees 268 Pi NU GeO TEG BUY IG FATE 26. ).<50: 2c ccsee Avge cascedecscpes cab) chs ec berteer Bede DpsersascoQineneapan vce 276 woumeerrcaware are, White-on-red style. .....ccscvcurerctissietiedesorscesbnsarsmunannstoosspahensuneescs 20 10. Cylindrical vessel, Interlocking fish patterm.....0....0.000000..cccccceceeeereeeseeees 278 Il. Cylindrical vessel, Interlocking fish pattern......0.00...00.0cccccceccccsteetecceeceeeees 279 12) armncrical vessel, Interlocking styles. .is22 os ccsicsecelistcviedssecieesandorvegeaenccen 280 13. Cormmaricai vessel: Interlocking styles. Joi. csideecevilevcssevc td ehanpoecsscordaal ove cgentieents 281 14, ulmmurieal vessel: Interlocking style siicn ckct icon dies essesnties we oeacsies todo. 282 15. Incurved bowl, Interlocking fish pattern...................c.cccceccsceescsecseseetsseeieeees 283 16. iment ved: bow), Interlocking fish pattertyi..0.2.1. 8.0.00 cenieadeles ued 284 17. ineuved. bow). Interlocking-style:z. 364 es ea ess on sd ene 285 18. Prcenrnt. Intenocking At y le... 55: cr8 kesh soho ac nsaek Seti gis ee 286 19. ererdsancerlocking 11aN Pattern ...ccc. accaes nec: )or eee een) ee ee 287 20. Paestern-on sherd, Interlocking styles: .itsen..cc. hs. ees Ah ccc et eco 288 PAS Darce sherd:. Interlocking stylenc ee eet een oy ena 288 22. Jar neck, Interlocking fish pattern........:....2:.:ce::cccusee Cage a ae 289 23. Ri amamiform jar FPOMy ete Boss cv... ee meeee eee ois Some ee eee ee eet 290 24-26. Jars and fragment attributed to Interlocking style... eee 290 266 University of California Publications in Am. Arch. and Ethn. [Vol. 21 SITES, GRAVES, AND STYLES Dr. Max Uhle’s collections at Chancay were made for the University of California in 1904. They comprise 701 catalogue entries, numbers 4—6361—-7051 and 4—9357-9366 of the University Museum of Anthro- pology. Of these, 531 are pottery vessels, mostly complete. These are here described and interpreted according to the plan followed in the monographs previously issued in this series on the pottery collections from Chincha, Iea, Ancon, Moche, and Supe.' The data filed by Dr. Uhle with reference to his work at Chaneay are contained in his field catalogue and in a general report of which the principal portions are herewith reproduced in the Appendix. He has also outlined and dis- eussed his results in an article, ‘‘Ueber die Friihkulturen in der Umegebung von Lima.’” Five sites were excavated by Dr. Uhle: Site A, La Mina, in a plain or gentle slope of soil at the northwestern foot of a rock or hill called Cerro de Trinidad. This hill is about 150 to 180 m. high and lies 400 m. south of the port of Chaneay, forming part of an elevation which separates the harbor and valley of Chancay from the salty meadows known as Las Salinas. Site B, La Calera de Lauren, 3 miles north of the town of Chaneay, on the sandy southern or seaward slope of a hill 200 m. high. There are ruins of tapia and a cemetery of burials in square pits, 1.5 to 3 m. deep, like those at site A. Textiles and perishable objects were not preserved. Site C, La Calera de Jegoan or Jecuan, on the landward side of the same hill. Site D, Huaral Viejo, Hacienda Guando. Site E, on the southern slope of the Cerro de Trinidad. Sites A and E are thus on opposite sides of a hill south of the harbor, B and C on opposite sides of a hill north of the harbor. As will be seen however, the ware from A and B (as well as D, which is not located with reference to the others) is identical, rather different from the ware of C, and quite different from that of E. The material at Site A was found in graves 1 to 5; that at B in graves 1 to 2, that at C in graves 0, 1-24, 26-30, 32-39. The vessels from D are from one grave. At site E numerous graves were exca- vated, but their contents were not designated separately. On the other hand Dr. Uhle separated the E ware into two groups of obviously 1 This volume, papers by Uhle, Strong, Kroeber. 2 Internat. Cong. Americanists, xv1, Vienna, 1908, 347-370, 1910. Cited here- after as Frihkulturen. . 1926 } Kroeber: The Uhle Pottery Collections from Chancay 267 different style, an E1 or ‘‘first period’’ and an E2 or ‘‘second period’’ ware. As he states in his appendix and in the article cited, the El ware was mostly broken and was found in the same graves with the whole E2 vessels. The provenience of objects according to site and grave is: Site A, grave 1, nos. 6361-6409; 2, 6410-20; 3, 6421-23; 4, 6424-28; 5, 6429-33. Site B, grave 1, 6434-49; 2, 6451-96. - Site C, grave 0, 6497-6501, 6518-20; 1, 6503-10; 2, 6511-14; 3, 6515-16, 6704; 4, 6521-22; 5, 6523; 6, 6524-25, 6684-86; 7, 6527-38; 8, 6540-43; 9, 6544; 10, 6545-49; 11, 6551-53; 12, 6554-55, 6687; 13, 6556-58; 14, 6559-67; 15, 6568-72; 16, 6573-75; 17, 6576-80; 18, 6581-85; 19, 6586-93; 20, 6594-6600; 21, 6601-03; 22, 6604-12; 23, 6613-20; 24, 6621-25; 26, 6626-29; 27, 6630-32; 28, 6633-34; 29, 6635; 30, 6636-39; ‘‘various,’’ ‘‘surface,’’ or ‘‘separate,’’ 6640-67, 6708; 32, 6668-73; 33, 6674-77; 34, 6678-81; 35, 6682; 36, 6683; 37, 6688-95; 38, 6696-99; 39, 6700-02. Site D, grave (1), 6705-26. Site E, ‘‘period 1,’’ 6727-6804, 6985, 7016, 7019; ‘‘period 2,’’ 6805-6984, 6986-7000, 7017-18, 7020; without specification, or ‘‘superficial,’’ 7001-15, 7021-51. Black-on-white graves, 9357-66. The styles represented at these five sites are five in number: A, B, D yielded only Black-on-white pottery of the style commonly known as that of Chaneay. This type of ware constitutes a consider- able part of the Ancon pottery which Strong has described as Late Ancon II. C yielded Red-white-black or Three-color Geometric ware; three and four-color base Epigonal; and a considerable proportion of the Black-on-white pottery which was found pure at A, B, and D. The E2 style is characterized by very simple white designs over- painted on red, and may be called White-on-red. Chancay. The El style may be described as three-color Interlocking, with crudely executed but rather intricate designs, and a high proportion of cylindrical vessels. BLACK-ON-WHITE STYLE: SITES A, B, D The familiar Black-on-white ‘‘Chaneay type’’ of ware was found by Dr. Uhle at three cemeteries without admixture of any other style. Since two of these cemeteries, A and B, were on the same two hills with cemeteries E and C that yielded wares of quite different style, a difference of period is rendered almost certain. The Black-on-white is light, thin, porous ware; its light red or orange-buff paste is rather crumbly; the white slip seales and decays 268 University of California Publications in Am. Arch. and Ethn. [Vol. 21 off easily. The black is often dark brown or even reddish brown. In a few large figure jars there is an overpaint of red on the lower part (pl. 806). Occasionally there is as much black area as white, or even more (pl. 81c), but in general the black forms slender designs. Smoked blackware (pl. 80a) is rare. Characteristic forms are: : 1. Ovoid jars, with a pair of handles either at the neck or at the middle of the body. The mouth is either concave or convex, and often ‘‘double’’ or ‘“swollen’’ (pls. 80e, 81f, g). Many of these jars are tall, rather narrow, and their bottoms almost pointed (pl. 80b-e). All the larger ones are somewhat flattened. The neck is sometimes modeled into a human face, characterized, like the Chancay figurines, by a sharp, narrow nose, usually by a chin that projects Figs. 1-4. Black-on-white style goblets and low bowls; 1, B2-6475; 2, B2-6485a; 3, B2-6480a; 4, B2-6480c. almost shelf-like, and by face painting that gives the effect of spectacles (pls. 80b, 8le, 82a). There is usually a knob, animal, or little platform modeled on the front below the neck (pls. 80c, d, 81g; cf. also 80b, 81a and the painting on 80e) or in smaller jars connected with one of the handles (pl. 80b, f). The painting extends over the upper two-thirds of the body of the jar. 2. Low open bowls, with or without a foot, more often with it. (pl. 82f, g; figs. 3, 4). 3. Incurved bowls (pl. 82c, e). 4. ‘*Quero’’-shaped unstemmed goblets or flat-bottomed unhandled cups. The profile is concavely conical (figs. 1, 2); some, with a foot, are almost hour- glass shaped. These goblets are usually plain white. These four types account for 107 of the 122 pieces from sites A, B, D. A few other forms appear in plates 80a (black), 80f (mammiform), 8ld (buff), 82b, 82d, and fig. 5 (mammiform). The patterns and design elements include: Simple narrow stripes (pl. 80d). Paired lines or stripes (pls. 80c, 82c). Stripes, paralleled by lines (pl. 82D). Same, with rows of dots (pls. 80c, 8la). On the lip of plate 80e the arrange- ment is of alternating triangles with dots. Stripes alternating with zigzag or wavy lines (pls 80d, 810, e, f). Pairs of lines diagonally crossed (pl. 81f, g). Toothed diagonals, triangles, or diamonds, or stripes broken by white squares (pls. 80c, d, 81a, g, 82e). 1926 } Kroeber: The Uhle Pottery Collections from Chancay 269 Small, free-standing animals or birds, curvilinear but conventionalized, usually repeated (pls. 80c, e, 82c), sometimes reduced to vestiges (pls. 80d, 82d). Low bowls mostly carry a pattern of two segments (pl. 82/, fig. 3); sometimes a pair of centripetal stripes or horn-like curves, inside or out (pl. 82/, fig. 4); or, the area between the two segments is patterned (pl. 829). A definite characteristic is bilateral asymmetry of design (pls. 80d, 81f/, 9), the two quadrant panels on each side of the median line containing dissimilar designs. A few jar paintings can be described as elaborate (pls. 80d, 81g), but most patterns are simple, and all rather hastily drawn. Dr. Uhle calls the Black-on-white ware late, and there is no reason to suppose otherwise. The site C pieces shown in plates 8le, g, 82d—g are Black-on-white in manner although found in graves also contain- ing Three-color Geometric and Epigonal ware. The two styles there- fore overlapped in time. The priority of Three-color and Epigonal over Black-on-white is indicated (1) by the relative antiquity of Three- color at Pachacamac*® and Moche,* where it is pre-Late Chimu and pre-Inca; (2) by the relations which Epigonal bears to Tiahuanaco ; (3) by the wide diffusion of Epigonal and Three-color (Pachacamac, Ancon, Supe, Moche), whereas Black-on-white is limited to the district from Huacho to Lima,® even Supe having revealed none, and (4) by Strong’s determination of Epigonal (Tiahuanaco) as occurring in Middle Ancon I and II, Three-color Geometric being characteristic of Late Ancon I, and Black-on-white of Late Ancon II. All this suggests Black-and-white as the latest ware of Chaneay and the immediately adjacent valleys. In that case one should expect to find it more or less associated with Inca and Late Chimu. Yet Dr. Uhle discovered no piece of either style in his three Chancay ceme- teries; and his Ancon excavations also yielded none, although Reiss and Stiibel found some Inca ware at Ancon.® The relative paucity of these late wares at and near Chancay (a complete absence is hardly prob- able) is the more remarkable in that Late Chimu is well represented, at least in variants, at Pachacamace,’ and its influence can be traced easily as far as Chincha® and Pisco,® while Inca is of course pan- Peruvian. There is not a single stirrup-mouth in the Black-on-white Chaneay collection, nor any of the modeling or relief ornamentation 3 Uhle, Pachacamae, pp. 35, 41, pl. 7, figs. 1-8, pl. 8, vs. pl. 13. 4This series, XxXI, pl. 62. 5 Am, Anthrop., u. s., in press, 1926. 6 See the classification in Strong, this volume, pp. 187-189. 7 Uhle, Pachacamac, pls. 13, 18. 8 This series, xxI, p. 14, fig. 4. 9 Specimens in the Peabody Museum of Harvard University and the Field Museum of Natural History. 270 University of California Publications in Am. Arch. and Ethn. [Vol. 21 characteristic of Late Chimu. The stippled piece d in plate 81 perhaps. shows a remote influence of Late Chimu relief stippling, but the piece is Central rather than North Peruvian in manner. It must be concluded then that the valley of Chancay, and in some measure the adjoining ones, were not seriously invaded by the late stylistic influences of the Chimus and the Incas. This facet would account for the development of a definite local style at a date when in the remainder of Peru these were being submerged, or at best were struggling against competition as at Ica. For this reason, per- haps, it came about that the late Chancay style attained to a two- color scheme. Color variety had evidently been shrinking in Peru for a considerable time until the Inea influence partly reinvigorated it; compare the superseding of red-and-white Proto-Chimu and three- color Pachacamac by prevailingly black Late Chimu, and the reduc- tion at Ica: Proto-Nazca 4-6 colors, Middle Ica 3-4, Late Iea 3 only, Inca 3-4. This general Peruvian tendency toward shrinkage of color scheme seems to have been carried to its undisturbed conclusion in the coast nook of Chaneay where for some reason it was not sub- jected to the color obliterating influences of Late Chimu or the color restoring ones of Inca. There are a few traits of Black-on-white which may be the result of indirect Inca influence or of influences affecting both. Such are: the frequent placing of handles low on the body of jars, the relatively sharp bottom, the knob or animal below the neck, the paneling of the design in quadrants. All these, however, are only remote suggestions of the aryballos: hints, if anything, that were remodeled to fit into a quite distinct and self-possessed style. Besides, the origin of the Cuzco style is not known. Dr. Uhle looks upon Chincha as having helped to form the Cuzco style. We need not go so far as to commit ourselves to this view, in the present state of imperfect knowledge, and yet must admit the definite possibility that the traits which Chincha, Cuzeo, and Black-on-white Chancay share are derived from a common though as yet undertermined source or set of influences, rather than that these styles are all of specific Cuzco origin. 1926] Kroeber: The Uhle Pottery Collections from Chancay to ~l THREE-COLOR AND EPIGONAL STYLES: SITE C The ware from site C, La Calera de Jecoan, is of several dis- tinguishable styles. It is well to state explicitly at the outset that these several styles occur again and again in association in the same grave.!° Many graves contain a preponderance of material in one style, and some contain one style exclusively. But these seem little more than normal variations of distribution, expectable from chance, especially with the number of vessels averaging less than 5 per grave." The Epigonal style was so named by Uhle on the ground of decadent derivation from the style of Tiahuanaco. It is well represented at Pachacamae, Ancon, Supe,'? and, in somewhat variant forms, as far north and south as Moche and Ieca.'* It seems however not to have been reported from the vicinity of Tiahuanaco itself; and Dr. J. C. Tello regards it as at least partly derivative from the old north Andean style of Chavin and Recuay and therefore only indirectly related to that of Tiahuanaco and perhaps its contemporary rather than suc- eessor.** Strong and Kroeber have also pointed out certain difficulties which the Uhle collections from Ica present toward the interpretation of Epigonal as derived from Tiahuanaco and have suggested the possibility of the reverse development. It must further be remem- bered that at Pachacamac, where Dr. Uhle first found and defined the Epigonal type, it occurs, according to his words, in the same graves with the rarer Tiahuanaco style.*® Whatever the origin and relations of the Epigonal style, its type is however clear, especially for the central Peruvian coast area. It is executed in 3 or 4 10 The following are the proportions in the graves with larger series of specimens: grave 0: black on white vessels 6, three color vessels 2; 1: 2, 4; Mee weed Or 7s Oo, 4) 83:0, 45 13: 1, 25 1455, 2: 155 '3,.22 17: 0, 45 18: 225 Pope eeecee, oy car 0, 15 2371; 4: 240 0-33 209 0, 37° S152, 1; 372e4, 1 ‘“Black on white’’ here includes all-white and red on white; ‘‘three color’’ includes four colors. The two groups correspond closely with the Black-on- white style on the one hand and the Epigonal and Three-color Geometric styles on the other. 11 One hundred eighty-one vessels in 38 graves. 12 Pachacamac: Uhle, pl. 5; Ancon: Strong, this volume, pl. 44; Supe: Kroeber, ibid., pl. 73. 13 Moche: Uhle, JSAP, figs. 16, 19, pl. 6, figs. 1-8, and Kroeber, this volume, pls. 63, 64, 66; Ica: Strong and Kroeber, ibid., pl. 30. 14 Dr. Uhle himself, in a recent letter, expresses the belief that part of the wares called Epigonal are not dependent on Tiahuanaco and perhaps anterior to it. 15 This volume, 118, 120. 16 Pachacamae, p. 22. 272 University of California Publications in Am. Arch. and Ethn. [Vol. 21 colors—in the latter case a deeper red being usually added to R, W, B17—is crudely painted in impure pigments, and without lustre. The style is most pronounced in cylindrical, flaring, and low goblets and in low bowls. Among its most pronounced features of design are small white rectangles (usually in rows) each containing a short bar; square faces, with or without feather head- dresses, and with the nose joined to the upper border; panels with rays or stripes or bars; pairs of half-interlocked open spirals or curves smooth or ser- rated; double or triple bars or step-pyramids projecting from the rim toward the center of bowls, usually from opposite sides; large dots or small circles, espe- cially in rows. These elements appear in plates 83b-i and 84c. Hpigonaloid are also the jars in plates 84g, h, 85b, the last having been found near the surface at site E. Epigonal is perhaps the least numerously represented style at site C. The Three-color Geometric style is characteristic of site C, without being numerically dominant. It has been described from Pachacamac, Ancon, and with some variation from Moche.'* It is not represented in the Uhle collection from Supe although that valley lies but little north of Chaneay. It is here named Three-color Geometric to dis- tinguish it from the Three-color Interlocking of site E and of the valley of Lima;'® and from the Three-color Textile, as the Late Chincha and Late Ica?® styles may be called. Three-color Geometric is characterized by its restriction to R, W, B, its overwhelming or exclusive use of gometric as opposd to representative orna- ment, a mediocre execution, and a dull finish. Characteristic of its designs are red stripes or broadish lines on a white ground, their angles filled with small black-bordered enclosures which often contain a dot or dash. The red ‘‘frame- work’’ is most typically a step, a zigzag, or a pair of zigzag lines crossing to form a row of diamonds. The little black-bordered outlines are, correspondingly, rectangles, isosceles triangles, and diamonds. Compare plates 83a, 84a, 85d, e, h, 1, which agree closely with the Pachacamac, Ancon, and Moche pieces already referred to. Related to the foregoing are red-white-black diagonally disposed squares, zigzag bands between stripes, zigzag bands containing S-scrolls, and smaller figures. Compare plates 83/, 84d-g, 85c, f, 1, several of which lean toward Epigonal. On the other hand, the jars in plate 85a, h, tend in the direction of Black- on-white in pattern and shape, though still three-color. 17 Nos. 6557, 6570, 6577, 6589, 6618, 6619, 6626, 6628 are four-color. The half dozen graves in which these four-color specimens were found, had Black- on-white vessels forming about one-third of their contents. No. 6570 is from grave 15, which held 3 B—W vessels out of 5; 6589 from 19, 7 B—W out of 8. The association of pure Black-on-white with the definite three and four-color Epigonal is thus certain. 18 Pachacamace: Uhle, pl. 7, especially figs. 1, 4, 5; pl. 8, figs. 2, 3, 4; Ancon:, Strong, this volume, pl. 43i-k (Late Ancon I); Moche: Kroeber, this volume pl. 62 (3 vessels from 2 graves only, but important for lying below a Late Chimu grave). 19 Uhle: Frihkulturen, especially figs. 4 (Chaneay), 5 (Pachacamac, but ‘¢Geometric’’ as much as ‘‘Interlocking), 10 (Chaneay), 16 (Aramburt). 20 Kroeber and Strong, this volume, pls. 11, 12, p. 17, fig. 6, 1924, and pls. 32-38, 1925 (‘‘Late Chincha I’’ and ‘‘Late Ica I,’’ also in part ‘‘ Late Chincha II’? and ‘‘ Middle Iea II’’ and ‘‘ Late Ica’’). : 1926] Kroeber: The Uhle Pottery Collections from Chancay 273 Black-on-white is the third style represented at site C. A number of vessels from this site have been included in the illustrations of Black- on-white ware (pls. 8le, g, 82d-g). In all of these except plate 82f some trace of Three-color Geometric manner is present. Black-on- white constitutes a good-sized minority of the vessels from C. As Dr. Uhle points out, the Calera de Jecuan series is interesting precisely because it comprises three styles—elsewhere found pure but here associated—plus transitions between them. It shows how the Black-on-white grew as a local specialization out of the more widely spread Three-color Geometric, and how this, in turn, links with the Epigonal. As Epigonal has 4 colors at times and represents heads and figures, it is more different from. Black-and-white and therefore pre- sumably at the opposite end of the temporal series from it; that is, earliest of the three. This statement refers to the time of origin and culmination of each style as a style: there is no intimation intended that the particular series of vessels from site C were anything else than contemporaneous. The overlapping in period of styles is a familiar phenomenon in Peru. That Epigonal is on the whole the earliest of the three styles here associated contemporaneously, accords well with the opinion generally held of it, whether that be Dr. Uhle’s view connecting it with the Tiahuanaco style or Dr. Tello’s deriving it from that of the northern highland. The site C Epigonal is a late Epigonal—faces and figures most rudimentary, execution slovenly. The Epigonal of Pachacamae and Supe, on the other hand, is less decayed; and it is also associated with ware in a style presumably originating earlier—more carefully drawn in detail, showing up to 5 and 6 colors, hard surfaced, and polished—the coast variety of Tiahuanaco.*t| We have then, for this central coast area, the time sequence Tiahuanaco—early Epigonal— late Epigonal—Three-color Geometric—Black-on-white, established both by the unmixed occurrence of some of the styles and by the association, with transitions, of each with the adjacent ones. Of bearing on the relation between Epigonal and Three-color Geometric, is the fact that the Epigonal style is most marked, on the whole, in goblets and low bowls, the Three-color in jars. So, at Supe, the goblets, bowls, and double-spouts show most of the Tiahuanaco and 21 Dr. Uhle calls it simply Tiahuanaco. But it is well to remember that in spite of its resemblances to the non-Inca ware from the Titicaca region, it differs from this. It has, for instance, forms apparently never reported from the Titicaca area—the double spout, bird and spout, jar with tapering face spout— besides numerous differences in designs. 274 University of California Publications in Am. Arch. and Ethn. [Vol. 21 Epigonal manner, the jars more modeling and simple painting. At Pachacamac, again, a comparison of Dr. Uhle’s plates 3, 4 with 7, 8 reveals the Tiahuanaco-Epigonal style most markedly in forms such as goblets, the Three-color in jars. In other words, Peruvian pottery painting styles show a tendency to erystallize in connection with cer- tain shapes on which they persist after other vessel forms have been introduced or have progressed to new shapes and ornamentation. This is a fact which must not be overlooked in the attempt to derive culture time-sequences from relations and associations of ceramic decoration. Rather remarkable is the practically total absence at Chancay of northern influences. There are very little modeling, almost no black- ware, and no Proto-Chimu or Late Chimu resemblances. The only stirrup-mouth is the hybrid form plate 85e, with wide flaring mouth and handles on the stirrup, painted in typical Three-color Geometric. It is a close counterpart of a Late Ancon I piece.” The bulk of the site C material is vessels of the same four shapes that constitute the overwhelming majority of the Black-on-white pieces from A, B, and D—jars, goblets, low and incurved bowls. In part this resemblance is due to the C series including Black-on-white specimens. In part, however, it is the result of the genetic relation of Three-color to Black-on-white. Form FREQUENCIES—PERCENTAGES Site C Sites A, B, D o) UPS Ny oh x es coe: op ae Un es 2 ol 49 Low BOWS ee ee ee ee 13 27 Ineurved t bowls 22.2 4284 3) eee 12 8 Gobletsiciccsccs es ook ee ee 12 4 Other formss.23332 eee 12 12 To halite eee a ee eee 100 100 Goblets have become less frequent in Black-on-white, bowls more frequent, jars remain constant. The following table shows more fully some of the variations between the lots, with the site C material subdivided according to color. It will be observed that the cylindrical goblet, high or low, has gone out, but the flaring goblet forms survive in Black-on-white; that the one-handled jar is almost extinct, but the unhandled figure jar has taken its place; and that handles are frequently set on the body in Black-on-white ware, more rarely in trichrome. The jar necks too are simpler in Three-color, as the plates show. The low bowl without the foot, and the incurved bowl with lip, seem typical of Black-on-white. 22 Mus. no. 4-5595, this volume, pl. 43n, 1925. 1926] Kroeber: The Uhle Pottery Collections from Chancay 270 ForM FREQUENCIES IN DETAIL Site C Site C Sites A, B, D Jars— R-W-B* B-Ww* B-W One handle—from neck .................. 12 Uf i; Two handles—from neck ................ 25 23 23 BP RUS ode nce cnc oop cceaceste go ani 6 ial 20 —rudimentary or lugs ...............- 5 3 6 Human figure, no handles .............. ee 1 93 10 60 Low Bowls— AWaine hoo bee eet ok fet iat 27 VALE OUDLMETOO ite ces eae ke eo te a ae 6 33 Incurved Bowls— A VEEN ULL a) geet ere ee rece a 4 13 7 QUOTED INU), Sa Lot ee oes 3) 22 2 9 Goblets— CENT OPN 0 cc Oe ee -2 ed re JEN RSTO TD: oot a ee nee 7 J 3 LO t: OLS pxeieeien see eset 2 ea 2 2 Doweeyind rica i...2 2.23 ea LeaN 7 ees 22 2 7 82 78 160 107 ie taeOiIn Gabe ee Noe oe 21 15 ANGI EEW TES = Pos (OE ee eee ey eee 181 122 WHITE-ON-RED AND INTERLOCKING STYLES: SITE E The site E material is perhaps the most interesting from Chaneay. That from the other sites has the value of relating known styles; that from E shows two new styles—new at least at the time of théir dis- covery. Owing to conditions encountered, Dr. Uhle did not inventory his vessels from site E by grave provenience. His most typical speci- mens in one of the two new styles, the actually interlocking ones, were mostly found broken and often incomplete in graves containing intact white-on-red ones. His argument, as set forth in the appendix, is that the Interlocking culture was the earlier, and that subsequent people of a lower culture, of which the White-on-red ware is repre- sentative, used the larger sherds of the earlier period as corpse covers. On this basis, he inventoried his E material as ‘‘first period’’ and ““second period.’’ However, he speaks of having succeeded in uncovering a few intact first period vessels, which had evidently been 23 R-W-B in this table includes R-B, R-W, R; B—W includes W. Roughly, R-W-B here means Three-color Geometric and Epigonal; B—W, Black-on-white style; but form and pattern do not always agree with color scheme. It will be recalled that a few good Black-on-white style pieces from A, B, D show some red. 276 University of California Publications in Am. Arch. and Ethn. [Vol. 21 found by the second period people in digging graves and had then been reinterred with their own dead. Actually the collection contains about four dozen whole vessels designated as ‘‘first period’’ plus a few similar ones of ‘‘second period’’ which are somewhat ecruder but essentially similar in type. Compare for instance plate 89f with 86f, and 90a with 86g. Plausible as Dr. Uhle’s interpretation of the situation at site E is, it is accordingly a subjective one. However, the objective facts as to association of specimens in the ground and within separate burials not having been obtainable or being no longer available, the collection will Figs. 5-7. White-on-red style jars; 5, C24—-6420 (found in association with Three-color and Epigonal ware, White-on-red in appearance, actually probably a base Epigonal specimen); 6, E-6833 (black, mammiform) ; 7, E-6986. be examined as divided by Dr. Uhle into lots El and E2. These two designations therefore do not, like Al or C36, refer to graves; nor are they employed with unreserved acceptance of Dr. Uhle’s view as to the lots being temporally distinct. They are used as enforced groupings which obviously conform in the main to a real distinction of some sort within the site E ware. THE WHITE-ON-RED STYLE E2 The E2 ware is the more numerous. It is simple in form and simple in painting. The ware is light red, strong, fairly thick and smooth, unstudied but not unpleasing in shape. About 30 per cent of it is unpainted; the remainder mostly has simple white designs over- painted on the red, occasionally black on white, or black and white on red, or all white. There is a low proportion of black vessels—not very successfully smoked. Nearly all the pieces look utilitarian, but scarcely any show fire-blackening or other signs of use. Modeling is almost lacking, and where attempted very inept. There is no clear resemblance to Inca, Tiahuanaco, Chimu, Nazea, or any of the better known Peruvian styles. 1926] Kroeber: The Uhle Pottery Collections from Chancay Pars The forms run wholly to bowls and jars, a full sixth of the collection being mammiform jars, large and small (pl. 86e; fig. 6). About half are ordinary jars, a third, bowls. Somewhat less than half the bowls have a lip (pls. 86, 87). The jars vary from almost cylindrical or globular forms without well defined neck (pl. 87f, 1, kK) to others with a cylindrical or even flaring mouth (pls. 86d, 879; fig. 8). Occasionally they are constricted one or more times, so as to resemble from two to four bowls set on top of one another (pl. 87a; fig. 9). These, with the crude pieces of plate 86f, g, a few vessels decorated with knobs or pro- jections (pls. 86e, 87g; fig. 8), and the breast forms represent the only attempts at non-utilitarian modeling. Not quite half the bowls and simple jars are handled. The bowls have the handles extending more or less horizontally, the jars usually vertically from the shoulder or neck; one-handled jars also occur (pl. 87f; fig. 7). There are no goblets and no cook pots; and none of the bowls has a foot, in distinction from the majority of low bowls from the Three-color and Black-on-white cemeteries. The subjoined list classifies the collection. Figs. 8, 9. Redware jars, White-on-red style; 8, E-6862; 9, E-6858. WHITE-ON-RED STYLE FORMS Bowls— PeGiGnnnet OW a Hari (Dl 81 G9) S02. 2. ate eens 14 Lipless, low, incurved or vertical-walled (87d, e) 16 Lipless, low, incurved, 2 handles (86a) —................ 6 36 War heel pena CUT V.CCi (SOG 1 1p.) esecnc sn oc eee cen ee 6 With lip, incurved, 2 handles (876) -..................... 18 24 60 Jars— . Broad mouth, no definite neck (87i, k)-...........-..-.-- 14 Broad mouth, 2-4 bulges (87a; fig. 9) -................... 4 18 Mortentor faring neck) \(860,.d) -s..ct 2 cabhe. 22 Vertical or flaring neck, 2 handles or knobs (fig.8) 25 Vertical neck, 1 flat or round handle on neck (87f; HL TC) EN aaa saa a meter segs, 2) 1 WR Re nig 58 IMCaAM TUE (OO sites Oy) eee rarer eee eee eee eee 30 30 Cylindrical, with or without rim of knobs (87g) -.. 3 Small mouth, large knobs or handles ..............-..--..--- 3 PROM G BDOUL COOT ) creed att caret aren een 2 Ey ror Be AECL pes le Rae VE ROR. Snaive Re es ers 3. FR LAs A 9 115 278 University of California Publications in Am. Arch. and Ethn. [Vol. 21 As for color, the tabulation that follows shows the prevalence of white design and red ground color, but also that this scheme is not exclusive, even three-color occurring. WHITE-ON-RED STYLE: COLOR Aetually white on yred tease eee eee ee 99 AT) Aya 6 exe oe eee an 5 Re aye ee tal eo ee ae 14 Black -onmewhite: 2.5.2 2. 408 ce eae te 5 Bisel: ands whitexomy rd aes ene ree ere ee ene 6 Smoked black: i.222e ee ee Cae ee ee 3 Plain®™ red ware’ 2222055, 3 ee eee eee 35 1 Any i S| SSAA EY \Sa ss cos San Sanu \S i Ny RY WS Kaw = RK mI | Seep doin Sgt . IN We Yuet 6 a — 0m, D Vege Lee Mitt100 Us, 75 NSS SNH O.3 Re LLILET Sieg, hl 5 gn. VW ‘4 Fig. 10. Cylindrical vessel, Interlocking fish pattern. R-W-B. E-6734. Designs are notably simple: dots, small circles, bars, angles, zigzag lines, diamonds, crossed lines. These are usually aggregated in from 2 to 5 rows or parallels. The dots and circles also come in clusters, follow lines, or fill spaces between them. The execution is as crude as the scheme is artless. A few patterns (pl. 87b, i) look as if they might be reductions from the triangular patterns common on the El Interlocking style vessels, but such interpretation must be advanced with reserve. Designs as simple as these White-on-red ones might be derived from almost any antecedents, and a linking would be legitimate only in the face of specific transitions. 1926 | Kroeber: The Uhle Pottery Collections from Chancay 279 THE INTERLOCKING STYLE E1 The lot of vessels designated as El by Dr. Uhle really comprises two or three groups which have little in common except the absence of the specific White-on-red characters of the E2 lot just discussed. Somewhat more than half of the series (a) consists of bowls and broad eylindrical jars with an interlocking fish, fret, or triangle pattern in ESSSSS99 » Kan LY ‘ Fa ; NE SSS NS AQAARIIES SS D Oo tee 5 Fig. 11. Cylindrical vessel, Interlocking fish pattern. R-W-B. E-6739. three colors. The smaller half of the series (b, c) varies greatly in form and color, shows as much modeling as painting, and connects with the interlocking style pieces chiefly, and somewhat dubiously, by the presence of several conventional fish designs, although these stand solitary and free. (a) Most of the true interlocking ware is broken, as stated by Dr. Uhle. The number of whole pieces, or such as can mainly be reassembled, is scarcely a dozen. About twice as many more are represented by sherds, some of them large, allowing the reconstruction of the pattern. This is always in three colors, 280 University of California Publications in Am. Arch. and Ethn. [Vol. 21 fundamentally R, W, B; but the red is sometimes pale, sometimes brownish, always dull, and the white runs often to yellow, buff, or gray, while some of the black is grayish.24 The pattern is therefore not salient, and in many eases is impossible to photograph. Its essential trait is an interlocking of the ele- ments, the engaging ones of which are in contrasting colors and between them » : N : $ CCE i Fig. 12. Cylindrical vessel, Interlocking style. R-W-B. E-6740. fill the decorative field. These elements are fishes, or possibly sometimes snakes, with triangular heads, and bodies bent on themselves and sometimes serrated (pl. 88c; figs. 10, 11,.15, 19, 22, perhaps 18). In borders there is an interlocking fret which seems to be a reduction of the same fish motive (pl. 88c, d,25 perhaps 89i; figs. 10-14, 16-17; fig. 19, similar but without inter- 24 The firing of the pottery was not infrequently uneven and unskilful, and several specimens have bulged or flattened during the process, as also in the H2 ware and that from site C. 25 Uhle, Frihkulturen, fig. 10. 1926] Kroeber: The Uhle Pottery Collections from Chancay 281 lock; fig. 20, with step). Perhaps related are triangles that suggest much reduced faces. These interlock, but do not contrast in color (pl. 89a, g; fig. 13). Other elements which are more or less worked into the interlocking scheme are zigzag lines (pls. 88d, 8971; figs. 12, 14) and rows of dots (pl. 88d; figs. 12, 14, 19); as the illustrations show, these tend to associate. These designs are dis- cussed in Dr. Uhle’s report on his collection, printed below in the Appendix. cA WSSESASAARRREEEREREEETTT SN K\ I . Fig. 13. Cylindrical vessel, Interlocking style. R-W-B. E-6748. Somewhat similar ware was found by Dr. Uhle at Aramburt in the valley of Lima,?¢ though there the fret and a step are more in evidence than the fish; and he adduces a fragment from Pachacamac, secured by him long after his classic excavations there.2’ As he also points out, the interlocking fish pattern is found in Proto-Nazcea; and it is from Proto-Nazca influence that he derives the present style.28 (b) A few jars seem related to the foregoing group through being painted with a serrated fish, usually single. One of these is a cylindrical jar like those 26 Frihkulturen, fig. 16. 27 Tbid., fig. 5. See also Uhle, Pachacamag, figs. 26-28. 28 Friihkulturen, p. 356 seq. The five vessels shown in fig. 7 are obviously from Nazca, as the text suggests, not from Chancay as the legend states. 282 University of California Publications in Am. Arch. and Ethn. [Vol. 21 of group (a); the others are mammiform, flat, or double-spouted (pls. 88b, 88e; fig. 21; also pls. 88a, 90d, whose form and texture ally them with the following group). (c) Finally, there is a varied assortment of pieces: double-spouts (pl. 89f, ef. 88a; fig. 26); bird or animal jars, poorly done (pl. 90a-c); human figure jars, rather mediocre in modeling (pl. 90e, f, h), and crude jars with heads (pls. 89e, 90g); blackware bowls (pl. 89b, d); a thin-spouted jar (fig. 24); a OTM TA AAA A ) SSS RRR S Aes NY Fig. 14. Cylindrical vessel, Interlocking style. R-W-B. E-6745. striped pitcher and jar (pl. 89e, h); a projection or knob of a very large jar (fig. 25); and a large flattish or mammiform jar with a design of a hexagonal face (pl. 88f). These are the most distinctive pieces. The face on the last mentioned relates to a face among the interlocking fish on plate 88c (fig. 10).29 Broad stripes appear in this group in plates 89e, f, h, 90b (ef. also fig. 24). The serrated fish of plates 88a and 90d has already been mentioned in connection with the preceding group. 29 Another face appears on the jar pl. 84b, which is ‘‘superficial’’ from E. The ‘‘frame’’ of this face suggests the serrated fish design; the ends of the serrations are Three-color Geometric; and the face is like Epigonal ones (pl. 83e). 1926] Kroeber: The Uhle Pottery Collections from Chancay 283 RELATIONS OF THE TWO STYLES AT E The material from site E is difficult to understand. The White-on- red, E2, is a definite style. So is the true Interlocking, Ela. That most of this was found fragmentary, whereas the White-on-red col- lection is prevailingly whole, goes to support Dr. Uhle’s explanation that people of one culture interred in the cemetery of another. How- ever, there are whole interlocking pieces; and there may have been many broken white-on-red ones.*° Lig L/P Fig. 15. Incurved bowl, Interlocking fish pattern. R-W-—B on unpainted ground. E-6746. A further complication is introduced by the heterogenous material which Dr. Uhle has allotted to his El period. The Elb group might be construed as still related to the interlocking Ela. The Elec lot can certainly not be so interpreted on the evidence of its own forms and designs. It is not only free from trace of interlocking patterns but quite variable inter se in every respect, even as regards texture. More- over, if Elb and Ele are classed with Ela into a single E1 style, the number of whole vessels in this style becomes too great to accord well with Dr. Uhle’s explanation that the E2 people encountered the El vessels in the ground and, purposely or in digging, broke them.*! 30 Dr. Uhle has paid more attention to sherds than most collectors in Peru; but he saved only those that seemed significant through a distinctive pattern. With nearly 200 entire vessels in hand, he would hardly have collected frag- ments of a ware so crude as White-on-red. Most of its fragments at that would be plain red and unmodeled: the sort. of sherds that occur at all Peruvian coast sites. 31 His published statement, Friihkulturen, p, 353, allows for more whole El vessels than his field report (Appendix, p. 297): ‘*‘Das Merkwiirdige.... ist, dass sich in ihren [E2] Grabern fast immer einzelne Gefasse, oder Reste 284 University of California Publications in Am. Arch. and Ethn. [Vol. 21 It remains to consider the affiliations of the styles. Dr. Uhle regards the White-on-red Chanecay style as related to the primitive or shellmound cultures of Ancon and Supe,*? but ‘‘a little different.’’** He holds that ‘‘the white painted rings and lines are a simple translation of previously engraved ornaments into painted ones, under the influence of their more advanced instructors [and predecessors, the El people].’’** This opinion seems venturesome. The step from incision to paint is not necessarily taken lightly by a people; and as to the designs themselves, the incised ones from Ancon and Supe have nothing actually in common with the painted ones from wd. a YA Fig. 16. Ineurved bowl, Interlocking fish pattern much reduced. R-B, inside unpainted. E-6781. * Chaneay except the comparative simplicity of both. Even the tech- nology, the color and texture, of the wares are considerably different, as are the forms. Nor can I see much relation between Ancon-Supe primitive ware and Proto-Nazea, which Dr. Uhle alleges.** It appears rather that after his discovery of Proto-Nazca in situ, he was so impressed with the antiquity of this style, that, not encountering it on the central coast, he equated with it, or rather derived from it, the simplest and presumably earliest culture which he found at Ancon and Supe. At Chanecay then, where the E1 Interlocking vessels do von einem hoch kultivierten Volke neben ihren eigenen ganz primitiven Topf- erein fanden .... Einzelne hervorragend schodne [El] Gefasse waren von ihnen [E2 people] aufgehoben, wahrscheinlich bentitzt und dann mit beigesetzt worden. ’’ 82 This volume, pls. 48, 79. 34 [bid. 33 Frihkulturen, pp. 352, 353. 35 Ibid., p. 356. 1926 | Kroeber: The Uhle Pottery Collections from Chancay 285 bear indubitable Proto-Nazca resemblances, he construed the associa- tion of these with the simple White-on-red as one of priority and sequence on the spot, and therefore derived the White-on-red from the incised Ancon-Supe and approximated it in time. This derivation and approximation perhaps influenced him to see a resemblance which is hard to discover.*° It is well to remember in regard to the primitive fishing or shell- mound culture of Ancon and Supe that the antiquity of this does not fly, ee W/, WL SELENA "Uitte... “Uf Y S FO AE Limi Mf, No, a 7 ”/ shah, lecrarhesstienrressenll Fig. 17. Incurved bowl fragment, Interlocking style. R-W-B. E-6803e. rest on stratigraphic evidence but on its being found unassociated with other ware, on its simplicity of style, and on its use of incising which is a rare Peruvian technique. Its chief claim to antiquity is its lack of clear relation to known Peruvian styles, not any determined rela- tion. This lack constitutes good presumption, but no proof of antiquity. As a matter of fact the co-occurrence in place and time of 36 The Uhle scheme seems to be: Earliest, Proto-Nazca. Next, derived from this, Proto-Lima, of which Chancay E1 Interlocking is a form or variant. Also influenced by Proto-Nazca, or related to it and therefore more or less contem- porary with it, is the primitive incised ware of the Ancon and Supe fishermen. The Chancay E2 White-on-red style is a development out of the incised Ancon- Supe style under some degree of influence of the El Interlocking. In Los Principios de las Antiguas Civilizaciones Peruanas, Bol. Soc. Ecuat. Estud. Hist. Am., Iv, no. 12, p. 11, Uhle makes the ancient fishing culture of Ancon con- temporary with the Proto-Nazca of Chincha and Pisco (est. ¢. 100 B.c.-50 A.c.), but continues early and later Proto-Nazca to ¢. 650 A.D. 286 University of California Publications in Am. Arch. and Ethn. [{ Vol. 21 distinct styles is so common in Peru** that the living side by side of strata or elements of population largely or wholly using ware of different styles must always be reckoned with as a possibility. i t ¥ . . a! 0 _ - . : * A To, gs ¥ * we a F al ¥ # a - - es re : = . ‘ 2 ; . ay ; - . % f Be . : ; : ; Paul Radin, ‘Pp. 1-160, Mca ‘Galifornia Culture Brey ies ‘by. ie September, 19 oY 2 aB., Winter and: Summer Danc Series te Ab - 17-216, 2 figures im text. August,’ 4 Ee of ‘the Pitch Indians, \e. Walla Lyons 7, (oni onieure aareeee™ in, h \\ Bp. 878-408, ‘plate 20, 182 fleur ie alifornian Kinship Terminologic : o with ¢ DS. Bere tor, 8, by Edward ‘Winslow itord Pp a8 A. wappo ‘Texts, First Series, by Paul Radin, Pp, ‘The. States eda ‘Hearst fee olnme as igur from pele ‘50-69, ed dances" in’ bides he ‘Uhle Pottery Collectio pier OL79) ' : _. Nos'5 and 6 in ont cove cations: of learand: societi and tnetit ns, wat e public ae the. cae vv e be