Ancient art of the province of Chiriqui, Colombia Sixth Annual Report of the Bureau of Ethnology to the Secretary of the Smithsonian Institution, 1884-1885, Government Printing Office, Washington, 1888, pages 3-188

Produced by Louise Hope, PM for Bureau of American Ethnology, The Internet Archive: American Libraries and the Online Distributed Proofreading Team at https://www.pgdp.net (This file was produced from images generously made available by the Bibliothèque nationale de France (BnF/Gallica) at http://gallica.bnf.fr)

[This text is intended for users whose text readers cannot use the "real" (Unicode/UTF-8) version. A few fractions such as 1/3 have been unpacked, and curly quotes and apostrophes have been replaced with the simpler "typewriter" form.

The fractions ½ and ¼ have been retained, as have a few accented letters and the "æ" combination.]

Ancient Art

of the

Province Of Chiriqui, Colombia.

by

William H. Holmes.

Contents.

Page. Introduction 13 Geography 13 Literature 14 Peoples 15 The cemeteries 16 The graves 17 Human remains 20 Placing of relics 21 Objects of art 21 Stone 21 Pictured rocks 21 Columns 22 Images 23 Mealing stones 25 Stools 27 Celts &c. 29 Spearheads 34 Arrowpoints 34 Ornaments 34 Metal 35 Gold and copper 35 Bronze 49 Clay: Pottery 53 Preliminary 53 How found 55 Material 55 Manufacture 56 Color 57 Use 57 Forms of vessels 58 Decoration 62 Unpainted ware 66 Terra cotta group 67 Black incised group 80 Painted ware 84 Scarified group 87 Handled group 90 Tripod group 97 Maroon group 107 Red line group 109 White line group 111 Lost color group 113 Alligator group 130 Polychrome group 140 Unclassified 147 Clay: Miscellaneous objects 149 Spindle whorls 149 Needlecases 150 Figurines 151 Stools 154 Musical instruments 156 Rattles 156 Drums 157 Wind instruments 160 Life forms in vase painting 171 Résumé 186 [Index]

Illustrations.

Page. PLATE I. Map of Chiriqui 13

Fig. 1. Section of oval grave 17 2. Section of a quadrangular grave 18 3. Grave with pillars 18 4. Compound cist 19 5. Southwest face of the pictured stone 22 6. A goddess of the ancient Chiriquians 23 7. A god of the ancient Chiriquians 24 8. Fragmentary human figure in gray basaltic rock 25 9. Mealing stone with large tablet ornamented with animal heads 26 10. Puma shaped metate 27 11. Stool shaped object 28 12. Stool with columnar base 28 13. Stool with perforated base 29 14. Large partially polished celt 30 15. Celt of hexagonal section 31 16. Small wide bladed celt 31 17. Celt with heavy shaft 31 18. Celt or ax with constriction near the top 31 19. Flaked and partially polished celt 32 20. Well polished celt 32 21. Narrow pointed celt 32 22. Narrow pointed celt 32 23. Cylindrical celt with narrow point 33 24. Leaf shaped objects suggesting spearpoints 34 25. Arrowpoints 34 26. Human figure, formed of copper-gold alloy 41 27. Grotesque human figure in gold 42 28. Rudely shaped human figure in gold 42 29. Grotesque human figure in nearly pure copper 43 30. Grotesque human figure in nearly pure gold 43 31. Rudely executed image of a bird in gold 44 32. Image of a bird in gold 45 33. Puma shaped figure in gold 45 34. Puma shaped figure in base metal 45 35. Quadruped with grotesque face in base metal 46 36. Figure of a fish in gold 46 37. Large figure of a frog, in base metal plated with gold 47 38. Small figure of a frog, in base metal plated with gold 47 39. Figure of an alligator in gold 48 40. Animal figure, in base metal plated with gold 48 41. Bronze bells plated or washed with gold 50 42. Bronze bell with human features 50 43. Triple bell or rattle found on the Rio Grande 51 44. Ancient Mexican bell 51 45. Fundamental forms of vases--convex outlines 58 46. Fundamental forms of vases--angular outlines 59 47. Vases of complex outlines--exceptional forms 59 48. Vases of compound forms 59 49. Square lipped vessel 59 50. Variations in the forms of necks and rims 60 51. Arrangement of handles 60 52. Types of annular bases or feet 61 53. Forms of legs 61 54. Grotesque figure forming the handle of a small vase 63 55. Grotesque figure forming the handle of a small vase 63 56. Grotesque figure forming the handle of a small vase 63 57. Monstrous figure with serpent shaped extremities 63 58. Monstrous figure with serpent shaped extremities 63 59. Grotesque figure 64 60. Grotesque figure 64 61. Grotesque figure 64 62. Figure of a monkey 64 63. Figure of a monkey 64 64. Figure of a monkey 64 65. Animal forms exhibiting long proboscis 65 66. Vase illustrating ornamental use of animal figures 65 67. Vase illustrating ornamental use of animal figures 65 68. Vase illustrating ornamental use of animal figures 66 69. Vase illustrating ornamental use of animal figures 66 70. Series of bowls and cups of unpainted ware 67 71. Vase of graceful form 68 72. Vase of graceful form 68 73. Vase of fine form, ornamented with grotesque heads 68 74. Vase of fine form, ornamented with grotesque heads 69 75. Vase with ornament of applied nodes and fillets 69 76. Vase with mantle covered with incised figures 70 77. Vase with frieze of grotesque heads 70 78. Vases with flaring rims and varied ornament 71 79. Vases with complex outlines and varied ornament 71 80. Large vase with two mouths and neatly decorated necks 72 81. Large vase with high handles 72 82. Top view of high handled vase 73 83. Handled vase 73 84. Handled vase 73 85. Handled vase 73 86. Small cup with single handle, ornamented with grotesque figure 74 87. Small cup with single handle, ornamented with grotesque figure 74 88. Vase of eccentric form 74 89. Vessel illustrating forms of legs 75 90. Vessel illustrating forms of legs 75 91. Vessel with large legs, decorated with stellar punctures 75 92. Vases of varied form with plain and animal shaped legs 75 93. Large vase of striking shape 76 94. Cup with legs imitating animal forms 76 95. Cup with legs imitating a grotesque animal form 77 96. Cup with legs imitating the armadillo 77 97. Cup with legs imitating the armadillo 77 98. Cup with frog shaped legs 77 99. Cup with legs imitating an animal and its young 77 100. Cups supported by grotesque heads 77 101. Large cup supported by two grotesque figures 78 102. Cup with two animal heads attached to the sides 78 103. Cup with two animal heads attached to the sides 78 104. Vase shaped to imitate an animal form 79 105. Vase shaped to imitate an animal form 79 106. Vase shaped to imitate an animal form 79 107. Fish shaped vessel 79 108. Top view of a fish shaped vessel 80 109. Cup with grotesque head attached to the rim 80 110. Black cup with incised reptilian figures 81 111. Black cup with incised reptilian figures 81 112. Black vase with conventional incised pattern 81 113. Small cup with conventional incised pattern 82 114. Small tripod cup with upright walls 82 115. Vase with flaring rim and legs imitating animal heads 82 116. Vase modeled to represent the head of an animal 83 117. Pattern upon the back of the vase 83 118. Tripod bowl of red scarified ware 87 119. Tripod bowl of red scarified ware 87 120. Oblong basin with scarified design 88 121. Large scarified bowl with handles imitating animal heads 88 122. Jar with flat bottom and vertical bands of incised ornament 89 123. Vase with stand and vertical incised bands 89 124. Vase with handles, legs, and vertical ribs 89 125. Tripod with owl-like heads at insertion of legs 90 126. Tripod with legs rudely suggesting animal forms 90 127. Heavy red vase with four mouths 90 128. Vase with horizontally placed handles and rude designs in red 91 129. Unpolished vase with heavy handles and coated with soot 92 130. Round bodied vase with unique handles and incised ornament 92 131. Vase with grotesque figures attached to the handles 93 132. Vase with upright handles and winged lip 93 133. Top view of vase with winged lip 94 134. Vase with grotesque animal shaped handles 94 135. Vase with handles representing strange animals 95 136. Vase with handles representing grotesque figures 95 137. Vase with handles representing animal heads 96 138. Vase with arched handles embellished with life forms in high relief 96 139. Vase with arched handles embellished with life forms in high relief 97 140. Tripod vase with shallow basin and eccentric handles 99 141. Tripod vase with shallow basin and eccentric handles 99 142. Tripod vase with shallow basin and eccentric handles 99 143. Tripod vase of graceful shape and neat finish 100 144. Heavy tripod vase with widely spreading feet 100 145. Neatly modeled vase embellished with life forms and devices in red 101 146. High tripod vase with incised designs and rude figures in red 101 147. Handsome tripod vase with scroll ornament 102 148. Vase with lizard shaped legs 102 149. Vase with scroll ornament 103 150. Large vase with flaring rim and widespreading legs 103 151. Fragment of a tripod vase embellished with figure of an alligator 104 152. Vase supported by grotesque human figures 105 153. Round bodied vase embellished with figures of monsters 106 154. Cup with incurved rim and life form ornamentation 107 155. Cup with widely expanded rim and constricted neck 107 156. Small tripod cup with animal features in high relief 108 157. Handsome vase supported by three grotesque figures 108 158. Vase decorated with figures of frogs and devices in red 110 159. Vase of unique shape and life form ornamentation 110 160. Two-handled vase with life form and linear decoration 110 161. Small tripod vase with animal figures in white 111 162. Shapely vase with designs in white paint 112 163. Small red bottle with horizontal bands of ornament 115 164. Small red bottle with encircling geometric devices 115 165. Bottle with zone occupied by geometric devices 116 166. Bottle with broad zone containing geometric figures 116 167. Bottle with decoration of meandered lines 117 168. Bottle with arched panels and geometric devices 117 169. Bottle with arched panels and elaborate devices 118 170. Vase with rosette-like panels 118 170a. Ornament from preceding vase 118 171. Vase with rosette-like panels 119 172. Vase with rosette-like panels 119 173. Theoretical origin of the arched panels 120 174. Theoretical origin of the arched panels 120 175. Theoretical origin of the arched panels 120 176. Vase decorated with conventional figures of alligators 120 177. Portion of decorated zone illustrating treatment of life forms 121 178. Portion of decorated zone illustrating treatment of life forms 121 179. Vase decorated with highly conventional life forms 121 179a. Design from preceding vase 122 180. Vase decorated with highly conventional life forms 122 181. Vase decorated with highly conventional life forms 123 182. Decorated panel with devices resembling vegetal growths 124 183. Vase of unusual shape 124 184. Vase of unusual shape 124 185. Vase of unusual shape 124 186. Double vessel with high arched handle 125 187. Double vessel with arched handle 125 188. Vase embellished with life forms in color and in relief 126 189. Vase modeled to represent a peccary 127 190. Under surface of peccary vase 127 191. Small vessel with human figures in high relief 127 192. Tripod cup with figures of the alligator 128 193. Large shallow tripod vase with geometric decoration 129 194. Large bottle shaped vase with high tripod and alligator design 130 195. Large bottle with narrow zone containing figures of the alligator 132 196. Vase with decorated zone containing four arched panels 133 197. Vase with four round nodes upon which are painted animal devices 133 198. Vases of varied form and decoration 134 199. Alligator vase with conventional markings 135 200. Alligator vase with figures of the alligator painted on the sides 135 201. Vase with serpent ornamentation 136 202. Vase representing a puma with alligator figures painted on sides 137 203. Shallow vase with reptilian features in relief and in color 137 204. Vase with funnel shaped mouth 138 205. Top view of vase in Fig. 204 139 206. End view of vase in Fig. 204 139 207. Large vase with decorations in red and black 140 208. Devices of the decorated zone of vase in Fig. 207, viewed from above 141 209. Handsome vase with four handles and decorations in black, red, and purple 142 210. Painted design of vase in Fig. 209, viewed from above 143 211. Vase of unusual shape with decoration in black, red, and purple 144 212. Ornament occupying the interior surface of the basin of vase in Fig. 211 144 213. Large vase of fine shape and simple decorations 145 214. Vase with extraordinary decorative designs 146 215. Painted design of vase in Fig. 214, viewed from above 147 216. Vase of unique form and decoration 148 217. Painted design of vase in Fig. 216 148 218. Spindle whorl with annular nodes 149 219. Spindle whorl decorated with animal figures 149 220. Spindle whorl with perforations and incised ornament 149 221. Needlecase 150 222. Needlecase 150 223. Needlecase with painted geometric ornament 151 224. Needlecase with incised geometric ornament 151 225. Needlecase with incised geometric ornament 151 226. Statuette 152 227. Statuette 152 228. Statuette 152 229. Statuette 152 230. Stool of plain terra cotta 154 281. Stool of plain clay, with grotesque figures 155 232. Stool of plain terra cotta 155 233. Rattle 157 234. Section of rattle 157 235. Rattle, with grotesque figures 157 236. Drum of gray unpainted clay 158 237. Drum with painted ornament 159 238. Painted design of drum in Fig. 237 159 239. Double whistle 161 240. Section of double whistle 161 241. Tubular instrument with two finger holes 162 242. Section of whistle 162 243. Small animal shaped whistle 162 244. Small animal shaped whistle 162 245. Top shaped whistle 163 246. Section, top, and bottom views of whistle 164 247. Drum shaped whistle 165 248. Vase shaped whistle 165 249. Crab shaped whistle 166 250. Alligator shaped whistle 166 251. Cat shaped whistle 167 252. Whistle with four ocelot-like heads 168 253. Bird shaped whistle 169 254. Bird shaped whistle 169 255. Bird shaped whistle 170 256. Whistle in grotesque life form 170 257. Conventional figure of the alligator 173 258. Conventional figure of the alligator 173 259. Conventional figure of the alligator 174 260. Conventional figure of the alligator 174 261. Conventional figure of the alligator 174 262. Conventional figure of the alligator 175 263. Conventional figure of the alligator 175 264. Conventional figure of the alligator 176 265. Conventional figure derived from the alligator 176 266. Conventional figure derived from the alligator 176 267. Conventional figure derived from the alligator 176 268. Conventional figure derived from the alligator 177 269. Conventional figure derived from the alligator 177 270. Conventional figure derived from the alligator 177 271. Conventional figure derived from the alligator 178 272. Conventional figure derived from the alligator 178 273. Conventional figure derived from the alligator 178 274. Conventional figures derived from the alligator 179 275. Conventional figure derived from the alligator 179 276. Conventional figure derived from the alligator 180 277. Conventional figures derived from the alligator 180 278. Conventional figures derived from the alligator 181 279. Conventional figures derived from the alligator 182 280. Conventional figures derived from the alligator 182 281. Conventional figures derived from the alligator 182 282. Conventional figures derived from the alligator 182 283. Conventional figures derived from the alligator 183 284. Vase with decorated zone containing remarkable devices 185 285. Series of devices 185

[Illustration: BUREAU OF ETHNOLOGY SIXTH ANNUAL REPORT PL. I.]

Ancient Art Of The Province Of Chiriqui.

By William H. Holmes.

Introduction.

Geography.

Until comparatively recent times the province of Chiriqui has remained almost unknown to the world at large. The isthmus was traversed a number of times by the conquerors, who published accounts of their discoveries, but it was reserved for the period of railroad and canal exploration to furnish trustworthy accounts of its character and inhabitants. The situation of Chiriqui is unique. Forming, politically, a part of South America, it belongs in reality to the North American continent. It occupies a part of the great southern flexure of the isthmus at a point where the shore lines begin finally to turn toward the north.

The map accompanying this paper (Plate I) conveys a clear idea of the position and the leading topographic features of the province. The boundaries separating it from Veragua on the east and Costa Rica on the west run nearly north and south. The Atlantic coast line has a northwest and southeast trend and is indented by the bay or lagoon of Chiriqui. The Bay of David extends into the land on the south and the Gulf of Dolce forms a part of the western boundary. A range of mountains, consisting principally of volcanic products, extends midway along the province, forming the continental watershed.[1] The drainage comprises two systems of short rivers that run, one to the north and the other to the south, into the opposing oceans. Belts of lowland border the shore lines. That on the south side is from twenty to thirty miles wide and rises gradually into a plateau two or three thousand feet in elevation, which is broken by hills and cut by cañons. This belt affords a natural thoroughfare for peoples migrating from continent to continent, and doubtless formed at all periods an attractive district for occupation. It is in the middle portion of this strip of lowland, especially in the drainage area of the Bay of David, that the most plentiful evidences of ancient occupation are found. Scattering remains have been discovered all along, however, connecting the art of Costa Rica with that of Veragua, Panama, and the South American continent. The islands of the coast furnish some fragmentary monuments and relics, and there is no doubt that a vast quantity of material yet remains within the province to reward the diligent search of future explorers.

[Footnote 1: For physical features, see report of Lieutenant Norton (Report Chiriqui Commission, Ex. Doc. 41, 1860).]

Literature.

The antiquarian literature of the province is extremely meager, being confined to brief sketches made by transient visitors or based for the most part upon the testimony of gold hunters and government explorers, who took but little note of the unpretentious relics of past ages. As there are few striking monuments, the attention of archæologists was not called to the history of primeval man in this region, and until recently the isthmus was supposed to have remained practically unoccupied by that group of cultured nations whose works in Peru and in Mexico excite the wonder of the world. But, little by little, it has been discovered that at some period of the past the province was thickly populated, and by races possessed of no mean culture.

The most important contributions to the literature of this region, so far as they have come to my knowledge, are the following: A paper by Mr. Merritt, published by the American Ethnological Society;[2] a paper by Bollaert, published by the same society, and also a volume issued in London;[3] a valuable pamphlet, with photographic illustrations, by M. De Zeltner, French consul to Panama in 1860;[4] a short paper by Mr. A. L. Pinart, published in the Bulletin de la Société de Géographie (Paris, 1885, p. 433), in which he gives valuable information in regard to the peoples, ancient and modern; and casual notes by a number of other writers, some of which will be referred to in the following pages. A pretty full list of authorities is given by Mr. H. H. Bancroft in his Native Races, Vol. V, p. 16.

One of the most important additions to our knowledge of the province and its archæologic treasures is furnished in the manuscript notes of Mr. J. A. McNiel, who made the greater part of the collection now deposited in the National Museum. This explorer has personally supervised the examination of many thousands of graves and has forwarded the bulk of his collections to the United States. His explorations have occupied a number of years, during which time he has undergone much privation and displayed great enthusiasm in pursuing the rather thorny pathways of scientific research. In the preparation of this paper his notes have been used as freely as their rather disconnected character warranted, and since Mr. McNiel's return to the United States, in July, 1886, I have been favored with a series of interviews with him, and by this means much important information has been obtained.

[Footnote 2: J. King Merritt: "Report on the huacals or ancient graveyards of Chiriqui." Bulletin of the American Ethnological Society, 1860.]

[Footnote 3: Bollaert: Antiquarian Researches in New Granada. London, 1860.]

[Footnote 4: A. De Zeltner: Notes sur les sépultures indiennes de département de Chiriqui.]

People.

At the present time this district is inhabited chiefly by Indians and natives of mixed, blood, who follow grazing and agriculture to a limited extent, but subsist largely upon the natural products of the country. These peoples are generally thought to have no knowledge or trustworthy tradition of the ancient inhabitants and are said to care nothing for the curious cemeteries among which they dwell, except as a source of revenue. Mr. A. L. Pinart states, however, that certain tribes on both sides of the continental divide have traditions pointing toward the ancient grave builders as their ancestors. There is probably no valid reason for assigning the remains of this region to a very high antiquity. The highest stage of culture here may have been either earlier or later than the period of highest civilization in Mexico and South America or contemporaneous with it. There is really no reason for supposing that the tribes who built these graves were not in possession of the country, or parts of it, at the time of the conquest. As to the affinities of the ancient middle isthmian tribes with the peoples north and south of them we can learn nothing positive from the evidences of their art. So far as the art of pottery has come within my observation, it appears to indicate a somewhat closer relationship with the ancient Costa Rican peoples than with those of continental South America; yet, in their burial customs, in the lack of enduring houses and temples, and in their use of gold, they were like the ancient peoples of middle and southern New Granada.[5]

The relics preserved in our museums would seem to indicate one principal period of occupation or culture only; but there has been no intelligent study of the contents of the soil in sections exposed in modern excavations, the exclusive aim of collectors having generally been to secure either gold or showy cabinet specimens. The relics of very primitive periods, if such are represented, have naturally passed unnoticed. Mr. McNiel mentions the occurrence of pottery in the soil in which the graves were dug, but, regarding it as identical with that contained in the graves, he neglected to preserve specimens.

In one instance, while on a visit to Los Remedios, a pueblo near the eastern frontier of Chiriqui, he observed a cultivated field about which a ditch some 8 or 9 feet in depth had been dug. In walking through this he found a continuous exposure of broken pottery and stone implements. Some large urns had been cut across or broken to conform to the slope of the ditch, and were exposed in section.

Although not apparently representing a very wide range of culture or distinctly separated periods of culture, the various groups of relics exhibit considerable diversity in conception and execution, attributable, no doubt, to variations in race and art inheritance.

[Footnote 5: R. B. White: Jour. Anthrop. Inst. Great Britain and Ireland, p. 241. February, 1884.]

The Cemeteries.

The ancient cemeteries, or huacals, as they are called throughout Spanish America, are scattered over the greater part of the Pacific slope of Chiriqui. It is said by some that they are rarely found in the immediate vicinity of the sea, but they occur in the river valleys, on the hills, the plateaus, the mountains, and in the deepest forests. They are very numerous, but generally of small extent. The largest described is said to cover an area of about twelve acres. They were probably located in the immediate vicinity of villages, traces of which, however, are not described by explorers; but there can be no doubt that diligent search will bring to light the sites of dwellings and towns. The absence of traces of houses or monuments indicates either that the architecture of this region was then, as now, of destructible material, or, which is not likely, that so many ages have passed over them that all traces of unburied art, wood, stone, or clay, have yielded to the "gnawing tooth of time."

One of the most circumstantial accounts of these burial places is given by Mr. Merritt, who was also the first to make them known to science.[6] Mr. Merritt was director of a gold mine in Veragua, and in the summer of 1859 spent several weeks in exploring the graves of Chiriqui; he therefore speaks from personal knowledge. In the autumn of 1858 two native farmers of the parish of Bugaba, or Bugava, discovered a golden image that had been exposed by the uprooting of a plant. They proceeded secretly to explore the graves, the existence of which had been known for years. In the following spring their operations became known to the people, and within a month more than a thousand persons were engaged in working these extraordinary gold mines. The fortunate discoverers succeeded in collecting about one hundred and thirty pounds weight of gold figures, most of which were more or less alloyed with copper. It is estimated that fifty thousand dollars' worth in all was collected from this cemetery, which embraced an area of twelve acres.

Although there are rarely surface indications to mark the position of the graves, long experience has rendered it comparatively easy to discover them. The grave hunter carries a light iron rod, which he runs into the ground, and thus, if any hard substance is present, discovers the existence of a burial. It is mentioned by one or two writers that the graves are in many cases marked by stones, either loose or set in the ground in rectangular and circular arrangements. The graves do not often seem to have had a uniform position in relation to one another or to the points of the compass. In some cases they are clustered about a central tomb, and then assume a somewhat radiate arrangement; again, according to Mr. McNiel, they are sometimes placed end to end, occupying long trenches.

[Footnote 6: J. King Merritt: Paper read before the American Ethnological Society, 1860.]

The Graves.

Graves of a particular form are said to occur sometimes in groups occupying distinct parts of the cemetery, but the observations are not sufficiently definite to be of value. The graves vary considerably in form, construction, and depth, and are classified variously by explorers. In the Bugaba cemetery Mr. Merritt found two well marked varieties, the oval and the quadrangular, reference being had to the horizontal section. The oval grave pits were from 4½ to 6 feet deep and from 3 to 4 feet in greatest diameter. A wall of rounded river stones 2½ to 3 feet high lined the lower part of the pit, and from the top of this the entire space was closely packed with rounded stones. Within the faced up part of this cist the remains of the dead, the golden figures, pottery, and implements had been deposited. This form is illustrated in Fig. 1 by a vertical section constructed from the description given by Mr. Merritt.

[Illustration: Fig. 1. Section of oval grave.]

The quadrangular graves were constructed in two somewhat distinct ways. One variety was identical in most respects with the oval form illustrated above. They were sometimes as much as 6 feet deep and frequently 4 by 7 feet in horizontal dimensions. In the other form a pit 4 by 6½ feet in diameter was sunk to the depth of about 3 feet. Underneath this another pit some 2 feet in depth was sunk, leaving an offset or terrace 8 or 10 inches in width all around. The smaller pit was lined with flat stones placed on edge. In this cist the human remains and the relics were placed and covered over with flat stones, which rested upon the terrace and prevented the superincumbent mass, which consisted of closely packed river stones, from crushing the contents. A section of this tomb is given in Fig. 2, also drawn from the description given by Mr. Merritt.

[Illustration: Fig. 2. Section of a quadrangular grave, showing the surface pack of river stories and the positions of the slabs and objects of art.]

[Illustration: Fig. 3. Grave with pillars, described by De Zeltner.]

[Illustration: Fig. 4. Compound cist, described by De Zeltner.]

Mr. Merritt and others mention that in some of the graves pillars are employed to support the roof of the cist. These pillars are mentioned briefly by De Zeltner, from whose account the following illustrations are drawn. This author does not state that he made any personal investigations, and if his accounts were obtained from the natives their entire trustworthiness may very properly be questioned. The first two forms mentioned by him are similar to those already given. The third is described as having at the corners square pillars of stone to support the covering, which, however, is not described. The fourth has four pillars, placed in the corners of the pit. These serve to support a vault of flagstones. The walls between the pillars are faced with pebbles, as in the cases previously described. Fig. 3 will make this form clear at a glance. The fifth variety described by De Zeltner is quite extraordinary in construction. His account is somewhat confusing in a number of respects, and the section given in Fig. 4 cannot claim more than approximate accuracy in details and measurements. Near the surface a paving, perhaps of river stones, was found covering an area of about 10 by 13 feet. This paving was apparently the surface of a pack about 2 feet thick, and covered the mouth of the main pit, which was some 6 or 7 feet deep. Pillars of cobble stones about 10 inches in diameter occupied the corners of the pit, and probably served in a measure to support the paving. In the bottom of this excavation a second pit was dug, the mouth of which was also covered by a paving 2½ by upwards of 3 feet in horizontal dimensions. This lower pit consisted of a shaft several feet in depth, by which descent was made into a chamber of inverted pyramidal shape. This chamber approximated 6 by 9 feet in horizontal dimensions and was some 4 or 5 feet deep. At the bottom of this cistern the human remains and most of the relics were deposited. The shaft was filled in with earth and the pavings described. The total depth, computed from the figures given, is about 18 feet, a most remarkable achievement for a barbarous people; yet this is equaled by the ancient tribes of the mainland of New Granada, where similar burial customs seem to have prevailed. Mr. White,[7] who traveled extensively in the northwestern part of the state, says:

A dry, elevated ridge, composed of easily excavated material, was selected as the cemetery. A pit of only a yard or so in diameter was sunk, sometimes vertically, sometimes at an angle, or sometimes it varied from vertical to inclined. It was sunk to depths varying from 15 to 60 feet, and at the bottom a chamber was formed in the earth. Here the dead was deposited, with his arms, tools, cooking utensils, ornaments, and chattels generally, with maize and fermented liquor made of maize. The chamber and passage were then rammed tightly full of earth, and sometimes it would appear that peculiar earth, other than that excavated on the spot, was used. One not unfrequently detects a peculiar aromatic smell in the earth, and fragments of charcoal are always found mixed with it in more or less quantity.

M. De Zeltner describes other very simple graves which are filled in with earth, excepting a surface paving of pebbles.

Mr. McNiel, who has examined more examples than any other white man, and over a wide district with David as a center, discredits the statements of De Zeltner in respect to the form illustrated in Fig. 4, and states that generally the graves do not differ greatly in shape and finish from the ordinary graves of to-day. He describes the pits as being oval and quadrangular and as having a depth ranging from a few feet to 18 feet. The paving or pack consists of earth and water worn stones, the latter pitched in without order and forming but a small percentage of the filling. He has never seen such stones used in facing the walls of the pit or in the construction of pillars. The flat stones which cover the cist are often 10 or 15 feet below the surface and are in some cases very heavy, weighing 300 pounds or more. A single stone is in cases large enough to cover the entire space, but more frequently two or more flat stones are laid side by side across the cavity. These are supported by river stones, a foot or more in length, set around the margin of the cist. He is of the opinion that both slabs and bowlders were in many cases carried long distances. No one of the pits examined was of the extraordinary form described in detail by De Zeltner and others.

[Footnote 7: B. B. White: Jour. Anthrop. Inst. Great Britain and Ireland, p. 246. February, 1884.]

Human Remains.

The almost total absence of human remains has frequently been remarked, and the theory is advanced that cremation must have been practiced. We have no evidence, however, of such a custom among the historic tribes of this region, and, besides, such elaborate tombs would hardly be constructed for the deposition of ashes. Yet, considering the depth of the graves, their remarkable construction, and the character of the soil selected for burial purposes, it is certainly wonderful that such meager traces of human remains are found. Pinart surmises, from the analogies of modern burial customs upon the north coast, that the bones only were deposited in the graves, the flesh having been allowed to decay by a long period of exposure in the open air. This, however, would probably not materially hasten the decay of the bones.

Mr. Merritt states that human hair was obtained from graves at Bugaba, and that he has himself secured the enamel of a molar tooth from that locality. De Zeltner tells us that in three varieties of graves remains of skeletons are found, always, however, in a very fragile condition. One skull was obtained of sufficient stability to be cast in plaster, but De Zeltner is not certain that it belonged to the people who built the tombs.

Mr. McNiel reports the occasional finding of bones, and a number of bundles of them are included in his collection. He reports that there are no crania and that nothing could be determined as to the position of the bodies when first buried.

Pinart observes that in some cases the bodies or remnants of bodies were distributed about the margin of the pit bottom, with the various utensils in the center, and again that the remains were laid away in niches dug in the sides of the main pit.

These scattering observations will serve to give a general idea of the modes of sepulture practiced in this region, but there must be a closer record of localities and a careful correlation of the varying phenomena of inhumation before either ethnology or archaeology can be greatly benefited.

Placing Of Relics.

The pieces of pottery, implements, and ornaments were probably buried with the dead, pretty much as are similar objects in other parts of America. The almost total disappearance of the human remains makes a determination of exact relative positions impossible. The universal testimony, however, is that all were not placed with the body, but that some were added as the grave was filled up, being placed in the crevices of the walls or pillars or thrown in upon the accumulating earth and pebbles of the surface pavement. The heavy implements of stone are rarely very far beneath the surface.

Objects Of Art.

From the foregoing account it is apparent that our knowledge of the art of ancient Chiriqui must for the present be derived almost entirely from the contents of the tombs. The inhabitants were skillful in the employment and the manipulation of stone, clay, gold, and copper; and the perfection of their work in these materials, taken in connection with the construction of their remarkable tombs, indicates a culture of long standing and a capacity of no mean order.

Of their architecture, agriculture, or textile art we can learn little or nothing.

The relics represented in the collection of the National Museum consist chiefly of articles of stone, gold, copper, and clay.

Stone.[8]

Works executed in stone, excluding the tombs, may be arranged in the following classes: Pictured rocks, sculptured columns, images, mealing stones, stools, celts, arrowpoints, spearpoints (?), polishing stones, and ornaments.

_Pictured rocks._--Our accounts of these objects are very meager. The only one definitely described is the "_piedra pintal_." A few of the figures engraved upon it are given by Seemann, from whom I quote the following paragraph:

At Caldera, a few leagues [north] from the town of David, lies a granite block known to the country people as the piedra pintal, or painted stone. It is 15 feet high, nearly 50 feet in circumference, and flat on the top. Every part, especially the eastern side, is covered with figures. One represents a radiant sun; it is followed by a series of heads, all, with some variation, scorpions and fantastic figures. The top and the other side have signs of a circular and oval form, crossed by lines. The sculpture is ascribed to the Dorachos (or Dorasques), but to what purpose the stone was applied no historical account or tradition reveals.[9]

[Illustration: Fig. 5. Southwest face of the pictured stone.]

These inscriptions are irregularly placed and much scattered. They are thought to have been originally nearly an inch deep, but in places are almost effaced by weathering, thus giving a suggestion of great antiquity. I have seen tracings of these figures made recently by Mr. A. L. Pinart which show decided differences in detail, and Mr. McNiel gives still another transcript. I present in Fig. 5 Mr. McNiel's sketch of the southwest face of the rock, as he has given considerably more detail than any other visitor. Mr. McNiel's sketches show seventeen figures on the opposite side of the rock. Seemann gives only twelve, while Mr. Pinart's tracings show upwards of forty upon the same face. These three copies would not be recognized as referring to the same original. That of Mr. Pinart seems to show the most careful study and is probably accurate. Good photographs would be of service in eliminating the inconvenient personal equation always present in the delineation of such subjects. These figures bear little resemblance to those painted upon the vases of this region.

Other figures are said to be engraved upon the bowlders and stones used in constructing the burial cists. De Zeltner states that "one often meets with stones covered with rude allegorical designs, representing men, pumas (tigre?), and birds. It is particularly in such huacas as have pillars and a vault that these curious specimens of Indian art are found."[10]

_Columns._--A number of authors speak casually of sculptured stone columns, none of which have been found in place. Seemann says that they may be seen in David, where they are used for building purposes,[11] but this is not confirmed by others. The sculptures are said to be in relief, like those of Yucatan and Peru. Cullen says that columns are found on the Island of Muerto, Bay of David.[12] Others are mentioned as having been seen in Veragua.

_Images._--Objects that may properly be classed as images or idols are of rather rare occurrence. Half a dozen specimens are found in the McNiel collections. The most important of these represents a full length female figure twenty-three inches in height. It is executed in the round, with considerable attempt at detail (Fig. 6). I may mention, as strong characteristics, the flattened crown, encircled by a narrow turban-like band, the rather angular face and prominent nose, and the formal pose of the arms and hands. Besides the head band, the only other suggestion of costume is a belt about the waist.

[Illustration: Fig. 6. A goddess of the ancient Chiriquians. Gray basalt--1/6.]

The material is a compact, slightly vesicular, olive gray, basaltic rock. I have seen a few additional examples of this figure, and from the identity in type and detail conclude that the personage represented was probably an important one in the mythology of the Chiriquians. In general style there is a rather close correspondence with the sculptures of the Central American States. Some of the plastic characters exhibited in this work appear also in the various objects of clay, gold, and copper described further on.

There is also a smaller, rudely carved, half length, human figure done in the same style. Besides these figures there are two large flattish stones, on one of which a rude image of a monkey has been picked, while the other exhibits the figure of a reptile resembling a lizard or a crocodile. The work is extremely rude and has the appearance of being unfinished. It seems that all of these objects were found upon the surface of the ground.

In Figs. 7 and 8 I present two specimens of sculpture also collected by Mr. McNiel, and now in the possession of Mr. J. B. Stearns, of Short Hills, N.J. The example shown in Fig. 7 was obtained near the Gulf of Dolce, 82° 55´ west. Three views are presented: profile, front, and back. It is carved from what appears to be a compact, grayish olive tufa or basalt, and represents a male personage, distinct in style from the female figure first presented. The head is rounded above, the arms are flattened against the sides, and the feet are folded in a novel position beneath the body. The height is 9 inches.

[Illustration: Fig. 7. A god of the ancient Chiriquians. Gray volcanic rock--½.]

The other specimen, Fig. 8, from near the same locality, is carved from a yellowish gray basalt which sparkles with numerous large crystals of hornblende. It is similar in style to the last, but more boldly sculptured, the features being prominent and the members of the body in higher relief. The legs are lost. Height, 5¼ inches.

A remarkable figure of large size now in the National Museum was obtained from the Island of Cana or Cano by Mr. McNiel. It is nearly three feet in height and very heavy. The face has been mutilated. In general style it corresponds more closely to the sculpture of the Central American States than to that of Chiriqui.

[Illustration: _a, b_ Fig. 8. Fragmentary human figure in gray basaltic rock--½.]

_Mealing stones._--The metate, or hand mill, which consists of a concave tablet and a rubbing stone, was an important adjunct to the household appliances of nearly all the more cultured American nations. It is found not only in those plain substantial forms most suitable for use in grinding grain, seeds, and spices by manual means, but in many cases it has been elaborated into a work of art which required long and skilled labor for its production.

In the province of Chiriqui these mills must have been numerous; but, since they are still in demand by the inhabitants of the region, many of the ancient specimens have been destroyed by use. It seems from all accounts that they were not very generally buried with the dead, but were left upon or near the surface of the ground, and were hence accessible to the modern tribes, who found it much easier to transport them to their homes than to make new ones.

The metates of Chiriqui present a great diversity of form and possibly represent distinct peoples or different grades of culture. They are carved from volcanic rocks of a few closely related varieties, the texture of which is coarse and occasionally somewhat cellular, giving an uneven or pitted surface, well suited to the grinding of maize. Three classes, for convenience of description, may be distinguished, although certain characters are common to all and one form grades more or less completely into another. We have the plain slab or rudely hewn mass of rock, in the upper surface of which a shallow depression has been excavated; we have the carefully hewn oval slab supported by short legs of varied shape; and we have a large number of pieces elaborately sculptured in imitation of animal forms. The first variety is common to nearly all temperate and tropical America and does not require further attention here. The second variety exhibits considerable diversity in form. The tablet is oval, concave above, and of an even thickness. The periphery is often squared and is in many cases ornamented with carved figures, either geometric devices or rudely sculptured animal heads. The legs are generally three in number, but four is not unusual. They are mostly conical or cylindrical in shape and are rather short.

The finest example of the second class has an oval plate 37 inches in length, 29 in width, and 2 inches thick, which is nearly symmetrical and rather deeply concave above. The central portions of the basin are worn quite smooth. Near the ends, within the basin, two pairs of small animal-like figures are carved, and ranged about the lower margin of the periphery are eighty-seven neatly sculptured heads of animals. There are four short cylindrical legs. This superb piece of work is shown in Fig. 9.

[Illustration: Fig. 9. Mealing stone with large tablet ornamented with animal heads, from Gualaca--1/9.]

Examples of the third class are all carved to imitate the puma or ocelot. The whole creature is often elaborately worked out in the round from a single massive block of stone. The thin tablet representing the body rests upon four legs. The head, which projects from one end of the tablet, is generally rather conventional in style, but is sculptured with sufficient vigor to recall the original quite vividly. The tail appears at the other end and curves downward, connecting with one of the hind feet, probably for greater security against mutilation. The head, the margin of the body, and the exterior surfaces of the legs are elaborately decorated with tasteful carving. The figures are geometric, and refer, no doubt, to the markings of the animal's skin. Nearly identical specimens are obtained from Costa Rica and other parts of Central America.

A fine example of medium size is given in Fig. 10. The material is gray, minutely cellular, basaltic rock. The upper surface of the plate is polished by use. The entire length is 17 inches.

[Illustration: Fig. 10. Puma shaped metate of gray andesite, from Rio Joca--¼.]

The largest specimen in the McNiel collection is 2 feet long, 18 inches wide, and 12 inches high. A similar piece has been illustrated by De Zeltner.

The usual office of these metates is considered to be that of grinding corn, cocoa, and the like. The great elaboration observed in some examples suggests the idea that perhaps they were devoted exclusively to the preparation of material (meal or other substances) intended for sacred uses. A high degree of elaboration in art products results in many cases from their connection with superstitious usages.

Speculating upon the use of these objects, De Zeltner mentions a mortar "whose pestle was nothing but a round stone, which still shows traces of gold here and there. It was evidently with the help of this rude instrument that the Indians reduced the gold to powder before fusing it."[13]

The implement or pestle used in connection with these mealing tablets in crushing and grinding is often a simple river worn pebble, as mentioned above, but is more usually a cylindrical mass of volcanic rock, worked into nearly symmetric shape.

_Stools._--The stool-like appearance of some of the objects described as metates suggests the presentation in this place of a group of objects that must for the present be classed as stools or seats, although their true or entire function is unknown to me. They are distinguished from the mealing stones by their circular plate, their sharply defined, upright, marginal rim, and the absence of signs of use.

[Illustration: Fig. 11. Stool shaped object carved from gray, minutely cellular basalt--1/3.]

[Illustration: Fig. 12. Stool with columnar base, carved from gray basaltic rock--1/3.]

Two of these objects are from the vicinity of David. The largest and most interesting is illustrated in Fig. 11. It is carved from a piece of vesicular basaltic tufa and is in a perfect state of preservation. The height is 6 inches and the diameter of the top 10 inches, that of the base being a little less. The slightly concave upper surface is depressed about half an inch below the upright marginal band. The periphery is a little more than an inch in width and is decorated with a simple guilloche-like ornament in relief. The disk-like cap is connected by open lattice-like work with the ring which forms the base. The interior is neatly hollowed out. The open work of the sides consists of two elaborately carved figures of monkeys, alternating with two sections of trellis work, very neatly executed. The other specimen is somewhat less elaborate in its sculptured ornament.

Outlines of two additional examples of these objects are given in Figs. 12 and 13. The tablets are round, thick, and slightly concave above and are margined with rows of sculptured heads. The supporting column in the first is a plain shaft and the base is narrow and somewhat concave underneath. In the second the column is hollowed out and perforated.

[Illustration: Fig. 13. Stool with perforated base, carved from gray basaltic rock--1/3.]

As bearing upon the possible use of these specimens it should be noticed that similar stool-like objects are made of clay, the softness and fragility of which would render them unsuitable for use as mealing plates or mortars, and it would also appear that they are rather fragile for use as stools. I would suggest that they may have served as supports for articles such as vases or idols employed in religious rites, or possibly as altars for offerings.

_Celts._--The class of implements usually denominated celts is represented by several hundred specimens, nearly all of which are in a perfect state of preservation. They are thoroughly well made and beautifully finished, and leave the impression upon the mind that they must represent the very highest plane of Stone Age art.

Although varying widely in form and finish there is great homogeneity of characters, the marked family resemblance suggesting a single people and a single period or stage of culture. They are found in the cists along with other relics and are very generally distributed, a limited number, rarely more than three, being found in a single grave. They may be classified by shape into a number of groups, each of which, however, will be found to grade more or less completely into the others. They display all degrees of finish from the freshly flaked to the evenly picked and wholly polished surface. The edges or points of nearly all show the contour and polish that come from long though careful use. All are made of compact, dark, volcanic tufa that resembles very closely a fine grained slate. The following illustrations include all the more important types of form. There are but few specimens of very large size. That shown in Fig. 14 is 8¼ inches long, 4 inches wide, and seven-eighths of an inch thick. The blade is broad at the edge, rounded in outline, and well polished. The upper end terminates in a rather sharp point that shows the rough flaked surface of the original blocking out. The middle portion exhibits an evenly picked surface. The rock is a dark slaty looking tufa, the surface of which displays ring or rosette-like markings, reminding one of the polished surface of a section of fossil coral. These markings probably come from the decomposition of the mineral constituents of the rock.

[Illustration: Fig. 14. Large partially polished celt of mottled volcanic tufa--½.]

The implement given in Fig. 15 may be taken as a type of a large class of beautifully finished celts. It also is made of the dark tufa, very fine grained and compact, resembling slate. The beveled surfaces of the blade are well polished, the remainder of the surface being evenly picked. The hexagonal section is characteristic of the class, but it is not so decided in this as in some other pieces in which the whole surface is freshly ground.

The contraction of the lateral outline and the sudden expansion on reaching the cutting edge noticed in this specimen are more clearly marked in other examples. The small celt shown in Fig. 16 is narrow above and quite wide toward the edge. A wide, thick specimen is given in Fig. 17. A specimen quite exceptional in Chiriqui is shown in Fig. 18. Mr. McNiel states that in many years' exploration this is the only piece seen that exhibits the constriction of outline characteristic of grooved axes.

[Illustration: Fig. 15. Celt of hexagonal section made of dark compact tufa--½.]

[Illustration: Fig. 16. Small wide bladed celt made of dark tufa--½.]

[Illustration: Fig. 17. Celt with heavy shaft made of dark speckled tufa--½.]

[Illustration: Fig. 18. Celt or ax with constriction near the top.]

[Illustration: Fig. 19. Flaked and partially polished celt of dark tufa--½.]

[Illustration: Fig. 20. Well polished celt of dark tufa--½.]

Two superb implements are illustrated in Figs. 19 and 20, the one in the rough excepting at the cutting edge, where it is ground into the desired shape, and the other neatly polished over nearly the entire surface. The surfaces are somewhat whitened from decomposition, but within the rock is nearly black, and the eye could not distinguish it from a dark slate. The material is shown by microscopic test to be a volcanic tufa. These examples were evidently intended for more delicate work than the preceding. The shapes of the specimens illustrated in Figs. 21 and 22 indicate a still different use. The upper end of the implement is large and rough, as if intended to facilitate holding or hafting, while the shaft diminishes in size below, terminating in a narrow, symmetrical, highly polished edge, a shape well calculated to unite delicacy and strength. The highest mechanical skill could hardly give to stone shapes more perfectly adapted to the manipulation of stone, metal, or other hard or compact substances. The material is a very dark, compact, fine grained tufa.

[Illustration: Fig. 21. Narrow pointed celt of dark tufa--½.]

[Illustration: Fig. 22. Narrow pointed celt of dark tufa--½.]

[Illustration: Fig. 23. Cylindrical celt with narrow point, of dark tufa--½.]

[Illustration: Fig. 24. Leaf shaped objects suggesting spearpoints, of dark tufa--½.]

An additional example is given in Fig. 23. The shaft is cylindrical and terminates in a conical point at one end and in a very narrow, abrupt, cutting edge at the other. The whole surface is polished. The material is the same dark tufa.

The class of objects illustrated in this and the two preceding cuts comprises but a small percentage of the chisel-like implements.

_Spearheads (?)._--Another class of objects made of the same fine grained, slaty looking tufa is illustrated in Fig. 24. They resemble spearpoints, yet may have been devoted to a wholly different use. They are long, leaf-like flakes, triangular in section, slightly worked down by flaking, sharpened by grinding at the point, and slightly notched at the top, perhaps for hafting.

_Arrowpoints._--The unique character of the arrowpoints of Chiriqui is already known to archæologists. The most striking feature is the triangular section presented in nearly all cases and shown in the figures (Fig. 25). The workmanship is extremely rude. The material is generally a flinty jasper of reddish and yellowish hues. The number found is comparatively small. The specimens given are of average size.

[Illustration: Fig. 25. Arrowpoints of jasper--1/1.]

_Ornaments._--It would seem from a study of our collections that ornaments of stone were seldom used by the inhabitants of Chiriqui. There are a few medium sized beads of agate and one pendant of dark greenish stone rudely shaped to resemble a human head. Ornaments of gold and copper were evidently much preferred.

[Footnote 8: I am indebted to Mr. J. S. Diller, of the United States Geological Survey, for the determination of the species of stone in this series of objects.]

[Footnote 9: Seemann: Voy. Herald, Vol. I, p. 312.]

[Footnote 10: A. de Zeltner: Notes sur les sépultures indiennes du département de Chiriqui.]

[Footnote 11: Seemann: Voy. Herald, Vol. I, p. 313.]

[Footnote 12: Cullen's Darien, p. 38.]

[Footnote 13: A. De Zeltner: Notes sur les sépultures indiennes, p. 7.]

Metal.

Gold And Copper.

The Chiriquians, like many of their neighbors in the tropical portions of the American continent, were skilled in the working of metals. Gold, silver, copper, and tin--the last in alloys with copper forming bronze--are found in the graves. Gold is the most important, and is associated with all the others in alloys or as a surface coating. The inhabitants of the isthmus at the time of the discovery were rich in objects, chiefly ornaments, of this metal, and expeditions sent out under Balboa, Pizarro, and others plundered the natives without mercy. When the Indian village of Darien was captured by Balboa (1510) he obtained "plates of gold, such as they hang on their breasts and other parts, and other things, all of them amounting to ten thousand pesos of fine gold."[14] From an expedition to Nicaragua the same adventurers brought back to Panama the value of "112,524 pieces of eight in low gold, and 145 in pearls."[15] Early Spanish-American history abounds in stories of this kind. Among others we read that Columbus found the natives along the Atlantic coast of Chiriqui and Veragua so rich in objects of gold that he named the district _Castillo del Oro_. It is said that the illusory stories of an _El Dorado_ somewhere within the continent of South America arose from the lavish use of gold ornaments by the natives whom the Spaniards encountered, and that Costa Rica gets its name from the same circumstance. It is also recorded that the natives of various parts of Central and South America at the date of the conquest were in the habit of opening ancient graves for the purpose of securing mortuary trinkets. The whites have followed their example with the greatest eagerness. As far back as 1642 the Spaniards passed a law claiming all the gold found in the burial places of Spanish America,[16] the whole matter being treated merely as a means of revenue.

The objects of gold for which the tombs of Chiriqui are justly famous are generally believed to have been simple personal ornaments, the jewelry of the primeval inhabitants, although it is highly probable that many of the figures, at least as originally employed, had an emblematic meaning. They were doubtless at all times regarded as possessed of potent charms, and thus capable of protecting and forwarding the interests of their owners. They have been found in great numbers within the last twenty-five years, but for the most part, even at this late date, have been esteemed for their money value only. Very many specimens found their way to this country, where they were either sold for curiosities or, after waiting long for a purchaser, even in the very shadow of our museums, were consigned to the melting pot. Many stories bearing upon this point have been told me. A Washington jeweler is represented as having exhibited in his window on Pennsylvania avenue about the year 1860 a remarkable series of these trinkets, most of which were afterwards sent to New York to be melted. About the same period a gentleman on entering a shop in San Francisco was accosted by a stranger who had his pockets well filled with these curious relics and wished to dispose of them for cash. A number of my acquaintances have neat but grotesque examples of these little images of gold attached to their watch guards, thus approving the taste of our prehistoric countrymen and at the same time demonstrating the identity of ideas of personal embellishment in all times and with all peoples.

The ornaments are found only in a small percentage of the graves, those probably of persons sufficiently opulent to possess them in life; a majority of the graves contain none whatever. They are often found at the bottom of the pits, and probably in nearly the position occupied by them while still attached to the persons of the dead. It is said that occasionally they are found in niches at the sides of the graves, as if placed during the filling of the pit.

Strangely enough, the gold is very generally alloyed with copper, the composite metal ranging from pure gold to pure copper. A small percentage of silver is also present in some of the specimens examined, but this is probably a natural alloy. In a few cases very simple figures appear to have been shaped from nuggets or masses of the native metals; this, however, is not susceptible of proof. The work is very skillfully done, so that we find it difficult to ascertain the precise methods of manipulation. The general effect in the more pretentious pieces resembles that of our filigree work, in which the parts are produced by hammering and united by soldering; yet there are many evidences of casting, and these must be considered with care. As a rule simple figures and some portions of composite figures present very decided indications of having been cast in molds, yet no traces of these molds have come to light, and there are none of those characteristic markings which result from the use of composite or "piece" molds. Wire was extensively used in the formation of details of anatomy and embellishment, and its presence does not at first seem compatible with ordinary casting. This wire, or pseudo-wire it may be, is generally about one-twenty-fifth of an inch in diameter.

The manner in which the numerous parts or sections of complex figures are joined together is both interesting and perplexing. Evidences of the use of solder have been looked for in vain, and if such a medium was ever used it was identical in kind with the body of the object or so small in quantity as to escape detection. At the junction of the parts there are often decided indications of hammering, or at least of the strong pressure of an implement; but in pursuing the matter further we find a singular perfection in the joining, which amounts to a coalescence of the metals of the two parts concerned. There is no weakness or tendency to part along the contact surfaces, neither is there anything like the parting of parallel wires in coils or where a series of wires is joined side by side and carried through various convolutions. In a number of cases I made sections of coils and parts composed of a number of wires, in the hope of discovering evidences of the individuality of the strands, but the metal in the section is always homogeneous, breaking with a rough, granular fracture, and not more readily along apparent lines of junction than across them; and further, in studying in detail the surface of parts unpolished or protected from wear by handling, we find everywhere the granular and pitted unevenness characteristic of cast surfaces. This is true of the wire forms as well as of the massive parts, and, in addition to this, such defects occur in the wires as would hardly be possible if they were of wrought gold.

All points considered, I am inclined to believe that the objects were cast, and cast in their entirety. It is plain, however, that the original model was made up of separately constructed parts of wire or wirelike strands and of eccentric and often rather massive parts, and that all were set together by the assistance of pressure, the indications being that the material used was sufficiently plastic to be worked after the manner of clay, dough, or wax. In one case, for example, the body of a serpent, consisting of two wires neatly twisted together, is held in the hand of a grotesque figure. The hand consists of four fingers made by doubling together two short pieces of wire. The coil has been laid across the hand and pressed down into it until half buried, and the ends of the fingers are drawn up around it without any indication of hammer strokes. Indeed, the effect is just such as would have been produced if the artist had worked in wax. Again, in the modeling of the eyes we have a good illustration. The eye is a minute ball cleft across the entire diameter by a sharp implement, thus giving the effect of the parted lids. Now, if the material had been gold or copper, as in the specimens, the ball would have been separated into two parts or hemispheres, which would not exhibit any great distortion; but as we see them here the parts are flattened and much drawn out by the pressure of the cutting edge, just as if the material had been decidedly plastic.

It seems to me that the processes of manufacture must have been analogous to those employed by the more primitive metal workers of our own day. In Oriental countries delicate objects of bronze and other metals are made as follows: A model is constructed in some such material as wax or resin and over it are placed coatings of clay or other substance capable of standing great heat. These coatings, when sufficiently thickened and properly dried, form the mold, from which the original model is extracted by means of heat. The fused metal is afterwards poured in. As a matter of course, both the mold and the model are destroyed in each case, and exact duplications are not to be expected. Mr. George F. Kunz, of New York, with whom I have discussed this matter, states that he has seen live objects, such as insects, used as models in this way. Being coated with washes of clay or like substance until well protected and then heavily covered, they were placed in the furnace. The animal matter was thus reduced to ashes and extracted through small openings made for the purpose.

As bearing upon this subject it should be mentioned that occasionally small figures in a fine reddish resin are obtained from the graves of Chiriqui. They are identical in style of modeling with the objects of gold and copper obtained from the same source.

In discussing possible processes, Mr. William Hallock, of the division of chemistry and physics of the United States Geological Survey, suggested that if the various sections of a metal ornament were embedded in the surface of a mass of fire clay in their proper relations and contacts they could then be completely inclosed in the mass and subjected to heat until the metal melted and ran together. After cooling, the complete figure could be removed by breaking up the clay matrix. I imagine that in such work much difficulty would be experienced in securing proper contact and adjustment of parts of complex figures. It will likewise be observed that evidences of plasticity in the modeling material would not exist. I must not pass a suggestion of Nadaillac[17] which offers a possible solution of the problem of manipulation. Referring to a statement of the early Spanish explorers that smelting was unknown to the inhabitants of Peru, he states that it would be possible for a people in a low state of culture to discover that an amalgam of gold with mercury is quite plastic, and that after a figure is modeled in this composite metal the mercury may be dissipated by heat, leaving the form in gold, which then needs only to be polished. There is, however, no evidence whatever that these people had any knowledge of mercury.

There is no indication of carving or engraving in the Chiriquian work. In finishing, some of the extremities seem to have been shaped by hammering. This was a mere flattening out of the feet or parts of the accessories, which required no particular skill and could have been accomplished with comparatively rude stone hammers. It is a remarkable fact that many, if not most, of the objects appear to be either plated or washed with pure gold, the body or foundation being of base gold or of nearly pure copper. This fact, coupled with that of the association of objects of bronze with the relics, leads us to inquire carefully into the possibilities of European influence or agency. I observe that recent writers do not seem to have questioned the genuineness of the objects described by them, but that at the same time no mention is made of the plating or washing. This latter circumstance leads to the inference that pieces now in my possession exhibiting this phenomenon may have been tampered with by the whites. In this connection attention should be called to the fact that history is not silent on the matter of plating. The Indians of New Granada are said to have been not only marvelously skillful in the manipulation of metals, but, according to Bollaert, Acosta declares that these peoples had much _gilt_ copper, "and the copper was gilt by the use of the juice of a plant rubbed over it, then put into the fire, when it took the gold color."[18] Just what this means we cannot readily determine, but we safely conclude that, whatever the process hinted at in these words, a thin surface deposit of pure gold, or the close semblance of it, was actually obtained. It is not impossible that an acid may have been applied which tended to destroy the copper of the alloy, leaving a deposit of gold upon the surface, which could afterwards be burnished down.

It has been suggested to me that possibly the film of gold may in cases be the result of simple decay on the part of the copper of the alloy, the gold remaining as a shell upon the surface of the still undecayed portion of the composite metal; but the surface in such a case would not be burnished, whereas the show surfaces of the specimens recovered are in all cases neatly polished.

If we should conclude that the ancient Americans were probably able to secure in some such manner a thin film of gold, it still remains to inquire whether there may not have been some purely mechanical means of plating. In some of the Chiriquian specimens a foundation of very base metal appears to have been plated with heavy sheet gold, which as the copper decays comes off in flakes. Occasional pieces have a blistered look as a consequence. Were these people able with their rude appliances to beat gold into very thin leaves? and Had they discovered processes by which these could be applied to the surfaces of objects of metal? are questions that should probably be answered in the affirmative.

The flakes in some cases indicate a very great degree of thinness. Specimens of sheet gold ornaments found in the tombs are thicker, but are sufficiently thin to indicate that, if actually made by these people, almost any degree of thinness could be attained by them. It would probably not be difficult to apply thin sheet gold to the comparatively smooth surfaces of these ornaments and to fix it by burnishing.

Mr. Kunz suggests still another method by means of which plating could have been accomplished. If a figure in wax were coated with sheet gold and then incased in a clay matrix, the wax could be melted out, leaving the shell of gold within. The cavity could then be filled with alloy, the clay could be removed, and the gold, which would adhere to the metal, could then be properly burnished down.

It will be seen from this hasty review that, although we may conclude that casting and plating were certainly practiced by these peoples, we must remain in ignorance of the precise methods employed.

Referring to the question of the authenticity of the specimens themselves, I may note that observations bearing upon the actual discovery of particular specimens in the tombs are unfortunately lacking. Mr. McNiel acknowledges that with all his experience in the work of excavation no single piece has been taken from the ground with his own hands, and he cannot say that he ever witnessed the exhumation by others, although he has been present when they were brought up from the pits. Generally the workmen secrete them and afterwards offer them for sale. He has, however, no shadow of a doubt that all the pieces procured by him came from the graves as reported by his collectors. The question of the authenticity of the gilding will not be satisfactorily or finally settled until some responsible collector shall have taken the gilded objects with his own hands from their undisturbed places in tombs known to be of pre-Columbian construction.

There are many proofs, however, of the authenticity of the objects themselves. It is asserted by a number of early writers that the American natives were, on the arrival of the Spaniards, highly accomplished in metallurgy; that they worked with blowpipes and cast in molds; that the objects produced exhibited a high order of skill; and that the native talent was directed with unusual force and uniformity toward the imitation of life forms. It is said that the conquerors were "struck with wonder" at their skill in this last respect. And a strong argument in favor of the genuineness of these objects is found in the fact that it is not at all probable that rich alloys of gold would have been used by Europeans for the base or foundation when copper or bronze, or even lead, would have served as well. We also observe that there is absolutely no trace of peculiarly European material or methods of manipulation, a condition hardly possible if the extensive reproductions were made by the whites. Neither are there traces of European ideas embodied in the shapes or in the decoration of the objects--a circumstance that argues strongly in favor of native origin. An equally convincing argument is found in the fact that all the alloys liable to corrosion exhibit marked evidences of decay, as if for a long period subject to the destructive agents of the soil. In many cases the copper alloy base crumbles into black powder, leaving only the flakes of the plating. Lastly and most important, the strange creatures represented are in many cases identical with those embodied in clay and in stone, and for these latter works no one will for a moment claim a foreign derivation.

Considering all these arguments, I arrive at the conclusion that the ornaments are, in the main, genuine antiquities, and that, if any deception at all has been practiced, it is to be laid at the door of modern goldsmiths and speculators, who, according to Mr. McNiel, are known in a few cases to have "doctored" alloyed objects with washes of gold with the view of selling them as pure gold.

I present the following specimens with a reasonable degree of confidence that all, or nearly all, are of purely American fabrication, and I sincerely hope that at no distant day competent archæologists may have the opportunity of making personal observations of similar relics in place.

The objects consist to a great extent of representations of life forms, in many cases more fanciful than real and often extremely grotesque. They include the human figure and a great variety of birds and beasts indigenous to the country, in styles resembling work in clay and stone of the same region. My illustrations show the actual sizes of the objects.

[Illustration: Fig. 26. Human figure with ridged crown, formed of copper-gold alloy.]

_The human figure._--Statuettes of men and women and of a variety of anthropomorphic figures of all degrees of elaboration abound. Fig. 26 illustrates a plain, rude specimen belonging to the collection of J. B. Stearns. It was obtained by Mr. McNiel from near the south base of Mount Chiriqui. The body is solid and the surface is rough and pitted, as if from decay. In many respects it resembles the stone sculptures of the isthmus. The metal is nearly pure copper. A piece exhibiting more elaborate workmanship, illustrated by Bollaert,[19] is shown in Fig. 27. Another remarkable specimen is illustrated by De Zeltner, but the photograph published with his brochure is too indistinct to permit of satisfactory reproduction. He describes it in the following language:

The most curious piece in my collection is a gold figure of a man, 7 centimeters in height. The head is ornamented with a diadem terminated on each side with the head of a frog. The body is nude, except a girdle, also in the form of a plait, supporting a flat piece intended to cover the privates, and two round ornaments on each side. The arms are extended from the body; the well drawn hands hold, one of them a short, round club, the other a musical instrument, of which one end is in the mouth and the other forms an enlargement like that of a flute, made of human bone. It is not probable that this is a pipe. Both thighs have an enlargement, and the toes are not marked in this little figurine.[20]

[Illustration: Fig. 27. Grotesque human figure in gold, from Bollaert.]

[Illustration: Fig. 28. Rudely shaped and finished human figure in gold.]

In Fig. 28 we have a rather rudely made and finished piece collected by Mr. McNiel, and now owned by Mr. Stearns. It exhibits features corresponding with a number of those referred to by De Zeltner. The foundation is thin and is of base metal coated with pure gold. I present two additional examples of the human figure from the collection of Mr. Stearns. One of them (Fig. 29) is an interesting little statuette in dark copper that still retains traces of the former gilding of yellow gold. The crown is flat and is surrounded by a fillet of twisted wire. The face is grotesque, the nose being bulbous, the mouth large, and the lips protruding. The hands are represented as grasping cords of wire which connect the waist with the crown of the figure and seem to be intended for the bodies of serpents, the heads of which project from the sides of the headdress. Similar serpents project from the ankles. The feet are flattened out as if intended to be set in a crevice. The extremities--excepting the feet--and the ornaments are all formed of wire. The various parts of the figure have been modeled separately and set together while the material was in a plastic or semiplastic condition. This is clearly indicated by the sinking of one part into another at the points of contact.

[Illustration: Fig. 29. Grotesque human figure in nearly pure copper, partially coated with yellow gold.]

[Illustration: Fig. 30. Grotesque human figure in nearly pure gold.]

An excellent example of the more elaborate figures is shown in Fig. 30. It is of reddish gold, slightly alloyed apparently with copper, and has in finishing received a very thin wash or plating of yellow gold, which is worn off in exposed parts. The central feature of the rather complicated structure is a grotesque human figure, much like the preceding, and having counterparts in both clay and stone. The figure is backed up and strengthened by two curved and flattened bars of gold, one above and the other below, as seen in the cut. The figure is decked with and almost hidden by a profusion of curious details, executed for the most part in wire and representing serpents and birds. Three vulture-like heads project from the crown and overhang the face. Two serpents, the bodies of which are formed of plaited wire, issue from the mouth of the figure and are held about the neck by the hands. The heads of the serpents are formed of wire folded in triangular form and are supplied with double coils of wire at the sides, as if for ears, and with little balls of gold for eyes. Similar heads project from the sides of the head and from the feet of the image.

The peculiarities of construction are seen to good advantage in this specimen. The figure is made up of a great number of separate pieces, united apparently by pressure or by hammering while the material was somewhat plastic. Upwards of eighty pieces can be counted. The larger pieces, forming the body and limbs, are hollow or concave behind. Nearly all the subordinate parts are constructed of wire.

[Illustration: Fig. 31. Rudely executed image of a bird in gold.]

_The bird._--Images of birds are numerous and vary greatly in size and elaboration. They are usually represented with expanded wings and tails, the under side of the body being finished for show. The back is left concave and rough, as when cast, and is supplied with a ring for suspension or attachment, as seen in the profile view (Fig. 31). The owl, the eagle, the parrot, and various other birds are recognized, although determinations of varieties are not possible, as in many cases the forms are rude or greatly obscured by extraneous details. The example shown in Fig. 31 is of the simplest type and the rudest workmanship, and is apparently intended for some rapacious species, possibly a vulture. The body, wings, and tail are hammered quite thin and are left frayed and uneven on the edges. The material appears to be nearly pure copper plated with yellow gold. Specimens of this class are very numerous. One, presented in a publication of the Society of Northern Antiquaries, and now in the museum at Copenhagen, is thought to be intended for a fish hawk, as it carries a fish in its mouth. De Zeltner mentions a statuette in gold of a paroquet, whose head is ornamented with two winged tufts. Such a specimen may be seen in the collection of Mr. Stearns.

[Illustration: Fig. 32. Image of a bird in gold, from Bollaert.]

Fig. 32 is reproduced from Bollaert. It represents a parrot and is very elaborately worked.

[Illustration: Fig. 33. Puma shaped figure in gold.]

[Illustration: Fig. 34. Puma shaped figure in base metal.]

_The puma._--Representations of quadrupeds are common; a good example, copied from Bollaert, is given in Fig. 33. The animal intended is apparently a puma, a favorite subject with Chiriquian workers in clay and stone as well as in gold. The body is hollow and open beneath and the fore feet are finished with loops for suspension. A similar piece with head thrown back over the body is shown in Fig. 34. The metal in this case appears to be nearly pure copper.

_Grotesque figure._--Another piece collected by Mr. McNiel is outlined in Fig. 35. The metal is quite base and the surface has been coated with gold, which is now nearly all rubbed off. The shape is that of a quadruped. The face has a rather grotesque, not to say satanic, expression. The details are not unlike those of other examples previously given.

[Illustration: Fig. 35. Quadruped with grotesque face in base metal.]

[Illustration: Fig. 36. Figure of a fish in gold. From Harper's Weekly, 1859.]

_The fish._--The fish was a favorite subject with the ancient nations of South America, and is modeled in clay, woven into fabrics, and worked in metals with remarkable freedom. It was in great favor in Chiriqui and must have been of importance in the mythology of the country. It occurs most frequently in pottery, where it is executed in color and modeled in the round. The very grotesque specimen in gold shown in Fig. 36 is copied from Harper's Weekly of August 6, 1859, where it forms one of a number of illustrations of these curious ornaments. The paper is, I believe, by Dr. F. M. Otis, who had just returned from Panama. A very curious piece owned by Mrs. Philip Phillips, of Washington, represents a creature having some analogies with the fish figure of Otis. Issuing from the mouth is the same forked tongue, each part terminating in a serpent's head. The body is about two inches long and the back has five triangular perforations. The tail is forked and the four leg-like members terminate in conventional serpents' heads. The metal is pure or nearly pure gold.

[Illustration: Fig. 37. Large figure of a frog in base metal plated with gold.]

[Illustration: Fig. 38. Small figure of a frog, in base metal plated with gold.]

_The frog._--The frog appears in the plastic art of Chiriqui more frequently perhaps than any other reptile. Its form is reproduced with much spirit and in greatly varying sizes, degrees of elaboration, and styles of presentation. It is probable that a number of species are represented. In Fig. 37 we have a large, rather plain specimen, now in the National Museum. The body and limbs are concave beneath, the metal being about one-sixteenth of an inch thick. Teeth are suggested by a number of perforations encircling the jaws and the eyes are minute hawk bells containing pellets of metal. The legs are placed in characteristic positions, and the hind feet are broad plates without indications of toes, a characteristic of these golden frogs. The framework or foundation is of copper, apparently nearly pure, and the surface is plated with thin sheet gold, which tends to flake off as the copper foundation corrodes.

The minute, delicately finished example given in Fig. 38 contrasts strongly with the preceding. It is also of base metal plated with pure gold and belongs to the collection of Mr. Stearns.

[Illustration: Fig. 39. Figure of an alligator, in gold, published in Harper's Weekly, 1859.]

_The alligator._--The alligator, which appears so frequently in the pottery of Chiriqui, is only occasionally found in gold. A striking specimen, illustrated in Harper's Weekly of August 6, 1859, is given in Fig. 39. A similar piece, formed of base metal, is in the collection of Mr. Stearns.

[Illustration: Fig. 40. Animal figure, in base metal plated with gold.]

_The crayfish (?)._--In Fig. 40 we have a fine specimen, intended apparently to represent a crayfish or some similar crustacean form. The head is supplied with complicated yet graceful antenna-like appendages, made of wire neatly coiled and welded together by pressure or hammering. The eyes are globular and are encircled by the ends of a double loop of wire which extends along the back and incloses a line of minute balls or nodes. The peculiar wings and tail will be best understood by referring to the illustration. The foundation metal is much corroded, being dark and rotten, and the plating of reddish gold seems to have been coated with a thin film of yellow gold. The profile view gives a good idea of the thickness of the metal and of the relief of the parts. Two rings or loops of doubled wire are attached to the extreme end of the nose and a heavy ring for suspending is fixed to the under side of the head.

_Miscellaneous._--Gold, pure and in the usual alloys, was also used in the manufacture of other articles, such as bells, beads, disks, balls, rings, whistles, thimble shaped objects, and amulets of varied shapes. Bells are more generally made of bronze, because, perhaps, of its greater degree of resonance. Thin plates, or rather circular sheets, of gold leaf are numerous. One mentioned by Bollaert was 7¼ inches in diameter. They are plain or crimped about the margins, indented in various ways, and sometimes perforated, apparently for suspension or attachment. Merritt mentions examples having holes which showed evidences of wear upon one side only, indicating attachment in a fixed position to some object or to some part of the costume. But one example is at hand, a thin sheet, three inches in diameter and crimped or indented neatly about the margin. Its thickness is about that of ordinary tinfoil.

[Footnote 14: Herrera: Hist. America, Vol. VI, p. 369.]

[Footnote 15: Herrera: Hist. America, Vol. III, p. 287.]

[Footnote 16: Mr. Hawes's letter answering questions about Chiriqui, read by Mr. Davis before the American Ethnological Society, April 17, 1860.]

[Footnote 17: Nadaillac: Prehistoric America, p. 450.]

[Footnote 18: Bollaert: Ethnological and Other Researches in New Granada, &c.]

[Footnote 19: Bollaert: Antiquarian Researches in New Granada, plate facing p. 31.]

[Footnote 20: A. De Zeltner: Notes sur les sépultures indiennes du département de Chiriqui.]

Bronze.

_Bells._--Bells seem to have been in pretty general use by the more cultured American races previous to the conquest. The form best known is the hawk bell, or common sleighbell of the North. The globular body is suspended by a loop at the top and is slit on the under side, so that the tinkling of the small free pellets of metal may be audible. Such bells are found in considerable numbers in the graves of Chiriqui, although I have no positive assurance that any of the examples in my possession were actually taken from graves which contained typical Chiriquian relics of other classes. The specimens now in the National Museum (Fig. 41) are in most cases, if not in all, of bronze, as determined by Mr. R. B. Riggs, of the chemical laboratory of the United States Geological Survey. All have been cast in molds. In most cases there are traces of a plating of gold. The largest is 1¼ inches in height and three-fourths of an inch in diameter. It is surmounted by the rude figure of an animal, through or beneath the body of which is an opening for the attachment of a cord. Others have simple loops at the top. The small perforated specimen belongs to Mr. Stearns. The additional piece given in Fig. 42 is unique in conception. It represents a human head, which takes an inverted position when the bell is suspended. The lower part of the bell forms a conical crown to the head and the ring of suspension is attached to the chin. Double coils of wire take the place of the ears, and the other features are formed by setting on bits of the material used in modeling. This specimen belongs to the collection of Mr. Stearns. Many examples of more elaborate workmanship have been recovered from the tombs and are now to be found in the collections of America and Europe.

[Illustration: Fig. 41. Bronze bells plated or washed with gold.]

[Illustration: Fig. 42. Bronze bell with human features.]

A specimen found many years ago on the Rio Grande, near Panama, and figured in Harper's Weekly, was of gold and showed specific variations from the Chiriquian pieces. It will be seen by reference to the outline given in Fig. 43 that three very neatly shaped and gracefully ornamented bells are mounted upon a circular plate to which a short handle is attached. It was evidently not intended for suspension, but rather to be held in the hand as a rattle.

A question as to the authenticity of these bells as aboriginal works very naturally arises, and it may be difficult to show to the satisfaction of the skeptical mind that any particular specimen is not of European origin or inspiration. At the same time we are not without strong evidences that such bells were in use by the Americans before the advent of the whites. Historical accounts are not wanting, but I shall only stop to point out some of the internal evidences of the native art. The strongest argument is to be found in the presence of analogous features in other branches of the art and in other arts. The eyes of the golden figures of reptiles are in many cases minute hawk bells, and in works of clay, the purely aboriginal character of which has not been called in question, similar features are discovered. The American origin of the bell, therefore, is not to be questioned. The form originated, no doubt, in the rattle, at first a nutshell or a gourd; later it was modeled in clay, and in time the same idea was worked out in the legs and the ornaments of vessels and in the heads and other parts of animal forms, which were made hollow and supplied with tinkling pellets. With the acknowledged skill of these people in the working of metals, there is no reason why the bells described should not have been manufactured independently of European aid and influence, provided the requisite metal was at hand.

[Illustration: Fig. 43. Triple bell or rattle found on the Rio Grande.]

[Illustration: Fig. 44. Ancient Mexican bell.]

It should be observed that if these early American bells were copied from or based upon Spanish originals they would not probably vary greatly in type with the various sections from which they are recovered, but it is observed that marked and persistent differences do occur. The well known Mexican bell, an example of which is outlined in Fig. 44, although of bronze, is generically distinct in form and construction.

In a brief review I may recall the more salient points regarding the use of metals in ancient Chiriqui. Gold, silver, copper, and apparently tin are represented.

Gold and copper were very plentifully distributed among the isthmian races, but we have little information as to the sources of supply. Free gold is found in the stream beds of many localities, and copper was probably found in its native state in some convenient locality; yet it is not impossible that these metals were transported from distant regions, as the inhabitants of Chiriqui must have had considerable intercourse with those of Central America on the north and with those of Granada on the south. Silver and tin are found in alloys with gold and copper, but not as independent metals. The silver gold alloy is probably a natural compound. In no case have I found silver to exceed 6 per cent. of the composite metal. Tin was artificially alloyed with copper, forming bronze. The latter metal resembles our ordinary bronze in color and hardness, but I am unable to secure more than a qualitative analysis on account of the scarcity of specimens available for the purpose. We have no information in regard to the origin of the tin. It is not found in a native state, and since it seems hardly probable that the Chiriquians understood smelting ores we are left in doubt as to whether it was obtained from more cultured nations to the north or to the south or from transoceanic countries.

The gold-copper alloys appear to range between pure gold and pure copper. If the bronze is of European origin, then we must conclude that all objects made of that metal are of post-Columbian manufacture. This question will probably be definitely settled in the near future.

The greater number of the objects were formed by casting in molds. Hammering was but little practiced, excepting, apparently, in the formation of sheet gold, which was probably an indigenous product. Repoussé work is not found, save as represented in the crimping and indenting of gold leaf. Engraving and carving were not practiced. It may be considered certain that gilding, or at least plating, was understood.

The objects are obtained from ancient graves of which no record or reliable tradition is preserved. They are all ornaments, no coin, weapon, tool, or utensil having come to my notice. The absence of utensils and of hammered objects of any kind strikes me as being rather extraordinary, since it is popularly supposed that, in the normal succession of events, hammering should precede casting and that utensils should be made before elaborate ornaments.

The work exhibits close analogies with that of the mainland of South America, but these analogies appear to be in material, treatment, and scope of employment rather than in the subject matter of the conceptions. The personages and zoömorphic characters represented are characteristically Chiriquian, and were derived no doubt from the mythology of the locality. These works affiliate with the various works in stone and clay, the art products of the province thus constituting a fairly homogeneous whole and being entirely free from traces of European influence.

Metals do not come into use early in the history of a race, as they are not found in shapes or conditions suitable for immediate use, nor are they sufficiently showy when found to be especially desirable for ornaments. A long period must have elapsed before the use of metals was discovered, and a longer period must have passed before they were worked; and, in the light of our knowledge of the ancient tribes of the United States, it would seem that a considerable degree of culture may be achieved before the casting of metals is understood; but in the ordinary course of progress the discovery of methods of alloying rare metals would be far separated from that of the simple fusing and casting of a single metal, such as gold. The Chiriquian peoples not only had a knowledge of the methods of alloying gold with copper, and, apparently, copper with tin, but, if our data are correct, they were able to plate the baser metals and alloys with sheet gold, and, what is far more wonderful, to wash them with gold, producing an effect identical with that of our galvanic processes.

The character of the conceptions embodied in the art unite with evidences of technical skill to prove to us that American culture, as represented by the metal ornaments of Chiriqui, was not the product of a day, but of long periods of experiment and progress.

Pottery.

_Preliminary._--The importance of the potter's art to archæology has often been pointed out. Baked clay is one of the most enduring materials utilized in art, and its employment by the races of men has fallen but little short of universal. The creations of that noblest of arts, architecture, and the antecedent forms of house building are necessarily left where erected, to be fed upon by the remorseless elements of nature, but the less pretentious utensil of clay accompanies its owner to the tomb, where it remains practically unchanged for ages.

Many glimpses of the early history of the American races and of the progress of art in pre-Columbian times are obtained through these exhumed relics, and in no case have we a view more clear and comprehensive than that furnished in the series here presented. The graves of Chiriqui have yielded to a single explorer upwards of 10,000 pieces of pottery, and this chiefly from an area perhaps not more than fifty miles square. These vessels constitute at least 90 per cent. of the known art of the ancient occupants of the province, and, although not so eloquent of the past as are the inscribed tablets of Assyria or the pictured vases of Greece, they tell a story of art and of peoples that without their aid would remain untold to the end of time.

A careful study of the earthenware of this province leads to the conclusion that for America it represents a very high stage of development, and its history is therefore full of interest to the student of art. Its advanced development as compared with other American fictile products is shown in the perfection of its technique, in the high specialization of form, and in its conventional use of a wide range of decorative motives. There is no family of American ware that bears evidence of higher skill in the manipulation of clay or that indicates a more subtile appreciation of beauty of form, and no other that presents so many marked analogies to the classic forms of the Mediterranean. Strangely enough, too, notwithstanding the well established fact that only primitive methods of manufacture were known, there is a parallelism with wheel made ware that cannot but strike the student with amazement.

In speaking thus of the whole body of ceramic products, I would not convey the impression that there is perfect homogeneity throughout, as if all were the work of a single people developed from within, and therefore free from the eccentricities that come from exotic influence. On the contrary, there is strong evidence of mixed conditions of races and of arts, the analysis of which, with our present imperfect data, will be extremely difficult. These evidences of mixed conditions are found in the marked diversity and individuality of character of the various groups of ware.

It is impossible, without the aid of careful observations in the field, to arrive at any conclusion as to the relative age of the different varieties of ware. Appearances of age are deceptive; the newer looking varieties may be the older and those executed in the most primitive style may belong to the later period, for grades in culture are not chronologic.

With reference to the principal groups of relics, we cannot do better than accept the statements of collectors that all are buried in like ways and in similar tombs, different varieties in many cases occurring in the same tomb. There are, however, in a few minor groups such marked distinctions in workmanship and style that we are compelled to attribute them to different periods or to distinct communities. The groups separated most completely from others are the scarified pottery presented first in the series of painted wares, the maroon group, which follows, and other varieties represented by fugitive pieces. The latter may have reached Chiriqui from neighboring provinces. There are certain pieces that speak decidedly of Costa Rican influence and others that find their counterparts in the Colombian states to the south.

In art in clay in most countries the vessel is the leading idea, the center about which nearly the entire ceramic art is gathered. This is true in a marked degree in Chiriqui, and vessels are therefore given the first place in this paper. The less usual forms include drums, whistles, rattles, stools, spindle whorls, needlecases, and toy-like images, all of which present features of peculiar interest. These classes of objects are discussed in separate sections.

There are few indications of an ambition to model natural forms or mythologic figures independently of utensils and useful objects, and, strange to say, no pieces are found that portray the human face and figure with even a fair degree of approach to nature.

_How found._--In describing the graves and tombs in a previous section, I alluded to the manner in which the pottery was deposited. It appears to have been buried with the dead or thrown into the grave with the earth and stones with which the pit was filled. There was little regularity in the place or position of the vessels and many were broken when found. The precise use of the vessels, the character of the contents, or the relation of particular pieces to the remains of the dead cannot be determined. Although the human remains have almost entirely disappeared and there are no traces left of utensils of wood, bone, horn, or shell, the paste, slip, and colors are wonderfully well preserved and the surface is not even discolored by contact with the earth. When found, every crevice and cavity is completely filled with earth, and the paste is often so tender that the vessels have to be dried with great care before they can be handled with freedom. The number of pieces found in a grave sometimes reaches twenty, but the average is perhaps not above three or four.

_Material._--The material used in the manufacture of this ware is remarkably uniform throughout the whole province, varying slightly with the locality, with the group, and with the character of the vessel constructed. Generally the paste consists of a matrix of fine clay tempered with finely pulverized sand, in which may be detected grains of quartz, feldspar, hornblende, augite, particles of iron oxide, &c. Argillaceous matter has been sparingly used, the sand in many cases comprising at least 75 per cent. of the mass. Many of the unpainted specimens, from which the polished slip has been removed, give off showers of fine sand when rubbed by the hand, and it is difficult to detect the presence of any finely comminuted matrix whatever. The thin slip employed in surface finish is more highly argillaceous than the paste. The clay used was probably mostly light in color, as the paste is now quite uniformly so. The baking was effected apparently without a very high degree of temperature and by methods that left few marks or discolorations upon the vessels. In hardness and durability the paste corresponds pretty closely with that of our red porous earthenware. The softer pieces can be scratched or even carved with a knife. Water will penetrate any of these vessels in a few minutes, but decay has probably tended to make the walls more porous.

_Manufacture._--There is no piece of this ware that does not bear evidence of a high degree of skill on the part of the potter; and yet, owing to the thorough manner in which the work is finished, the precise methods of manipulation are not easily detected. So great is the symmetry and so graceful are the shapes that one is led to suspect the employment of mechanical devices of a high order. The casual observer would at once arrive at the conclusion that the wheel or molds had been used, but it is impossible to detect the use of any such appliances. We observe that irregular and complex forms, in the production of which mechanical appliances could not be used to advantage, are modeled with as much grace of contour and perfection of surface as are the simpler shapes that could be turned upon a wheel, and we conclude that with this remarkable people the hand and the eye were so highly educated that mechanical aids were not indispensable. I find no evidence that coil building was systematically practiced, but it is clear that parts of complex forms were modeled separately and afterwards united. The various ornaments in relief (the heads and other parts of animals) and the handles, legs, and bases of vessels were constructed separately and then luted on, and with such skill that the thinnest walls and the most complex and delicate forms were not injured in the process. The contact irregularities were then worked down, and every part of the surface, including the more important ornaments, were rendered smooth, preparatory to the application of the thin surface wash or slip. After the slip was applied and the clay became somewhat indurated, the surface was polished with smooth pebbles, the marks of which can be seen on the less accessible parts of the vessel. On the exposed surfaces of certain groups of ware the polish is in many cases so perfect that casual observers and inexperienced persons take it for a glaze. Incised figures and painted decorations were generally executed after the polishing was complete. Details of processes will be given as the various classes of ware pass under review.

The methods of baking were apparently of a higher order than those practiced in many parts of America. One rarely discovers traces of the dark discolorations that result from primitive methods of baking, yet there are none of the contact marks that arise from the furnace firing of Spanish-American potters.

_Color._--The colors of the ware and of the surface applications vary decidedly with the different groups. The prevailing colors of the paste may be defined as ranging from very light yellow grays to a variety of ochery yellows and very pale terra cotta reds. In one or two groups there is an approach to salmon and orange hues, and in another the color is black or dark brown. The color within the mass is in some cases darker than upon the surface, an effect produced in baking, and not through the use of different clays. The slip is usually lighter than the surface of the paste.

The colors used in finishing and decorating are confined to reds, blacks, and purple grays. In one large group of ware the appearance of the delineations is such as to lead to the conclusion that the principal pigment or fluid employed in delineation has totally disappeared, carrying with it all underlying colors not of unusual permanence or not worked down with the polishing implement. The Aztec and other races of tropical America used an argillaceous, white pigment in decorating their wares, which has in many cases partially or wholly disappeared, carrying away considerable portions of the colors over which it was laid, while in other cases, and also in this Chiriqui ware, there is no trace of color remaining and we are left to surmise that the brush used probably contained merely a "taking out" medium. Red was profusely used and varies from a light vermilion to a deep maroon. In certain classes of vessels it was hastily daubed on, covering prominent parts of the surface or forming irregular spots, streaks, and rude figures. In two groups of ware it was used as the chief delineating color. In some cases it was employed as a wash or slip and was worked down with the polishing stone, and in this condition it was treated as a ground upon which to execute designs in other colors. It is always a fast color and is probably of mineral character.

The blacks are of two kinds, which are used in distinct groups of ware: one, probably a mineral pigment, somewhat pasty when applied and quite permanent, is always used in delineating the ornamental figures; the other, possibly a vegetable tint, is always used as a ground upon which to execute designs in other mediums. It is confined to a single group of ware. It has in many cases disappeared entirely, and where remaining can be removed with ease by rubbing.

A light purple tint is tastefully and sparingly employed in one group of ware. Browns and other hues occur but rarely and in all cases result from alterations of other colors produced in firing. The color effects of this pottery, although evidently much modified by age, are sufficiently rich to be highly pleasing to the eye.

_Use._--The uses to which most classes of earthen products were applied are easily determined. Whistles, drums, rattles, and spindle whorls have definite duties to perform, and vessels, as to general scope of function, answer for themselves: but when we come to inquire into the particular uses of the various groups of vessels we are often at a loss. The majority of the pieces show no abrasion by handling or discoloration by fire or by contents, and I am inclined to believe that a large portion were taken directly from the furnace and deposited in the tombs. This implies manufacture for purely mortuary purposes.

Two important groups, the high tripods and the two handled cups or pots, are generally discolored by use over fire, but we cannot say with confidence whether that use was a domestic one or whether it was ceremonial. The small size and the elaborate modeling of a majority of the pieces make it appear improbable that they were intended for use in ordinary cooking or even in the preparation of beverages. A few large plain caldrons are found, and these were probably domestic receptacles. All things considered, it would seem highly probable that the greater portion of the vessels exhumed from the graves were intended to be used for religious and mortuary purposes.

The preceding paragraphs refer, for the most part, to the whole body of earthenware products, but throughout the rest of this section I shall treat of vessels only, except in the matter of decoration, which refers equally to all classes of objects.

[Illustration: _a, b, c, d, e, f_ Fig. 45. Fundamental forms of vases--convex outlines.]

[Illustration: _a, b, c, d, e_ Fig. 46. Fundamental forms of vases--angular outlines.]

[Illustration: _a, b_ Fig. 47. Vases of complex outlines--exceptional forms.]

_Forms of vessels._--Divesting the utensil of extraneous features, such as rims, handles, and legs, we have the following series of shapes, which shows a pretty full graduation of outline from extreme to extreme. Beginning with the simplest fundamental form, the shallow cup (Fig. 45, _a_), we ascend gradually to more complex outlines, such as are seen in the hemispherical bowl (_b_), the deep basin with slightly incurved rim (_c_), the globular form (_d_), and the elongated form (_e_). Occasionally we see an eccentric variation, such as is shown in _f_. Flat bottoms are unusual; a conical base is the rule. Outlines do not always exhibit these even, convex curves, but many are straight or concave in profile, as shown in Fig. 46. Complex forms are shown in Fig. 47, _a_ and _b_, and compound forms in Fig. 48, _a_ and _b_. Examples of these classes are numerous and important. The compound shapes result from the union of two or more simple forms. Eccentric forms are numerous and result in a majority of cases from the employment of some animal as a model. Thus, if an alligator or almost any quadruped is embodied in the vessel, the form tends to become elongated; if a crab or a fish is imitated, there is a tendency to flatness &c. The base is almost universally more or less conical, is rarely flat, and never concave, excepting as the result of the addition of an annular foot or stand. The radical shapes do not undergo any considerable change when rims, necks, handles, legs, and other appendages are added. The rim or lip is in many cases incurved, but as a rule it is turned outward. The margin is plain, symmetrical, and often considerably thickened. In a few instances the outline is rectangular or scalloped, as shown in Fig. 49, and the attachment of handles often leads to peculiar outlines, as will be seen further on.

[Illustration: _a, b_ Fig. 48. Vases of compound forms.]

[Illustration: Fig. 49. Square lipped vessel.]

[Illustration: _a, b, c, d_ Fig. 50. Variations in the forms of necks and rims--various groups of ware.]

The neck in its simplest form is a narrow upright band surrounding the orifice (Fig. 50, _a_) and is not differentiated from the rim. Variations in size and shape are shown in the remaining figures of the series. In _b_ it is a narrow constricted band beneath an overhanging rim, in _c_ it is upright and considerably elongated, and in _d_ it expands, giving a funnel shaped mouth. The exterior surface is very generally decorated with relieved or painted devices. High necked bottles and pitcher shaped vessels are unknown.

[Illustration: _a, b, c, d, e, f_ Fig. 51. Arrangement of handles--various groups of ware.]

Handles constitute a very interesting feature of this pottery and are much varied in shape and arrangement. In a few cases the handle is a single arch springing over the orifice, as seen in Fig. 51, _a_. Again, the handle is attached to one side, as in _b_, but as a rule handles occur in twos upon the shoulder, one on either side of the aperture. They are horizontally attached, as in _c_, or vertically placed, as in _d_, connecting the rim with the shoulder, or they occur low on the body, as in _e_. In rare cases there are four handles, which are arranged as seen in _f_ or are set on in pairs. In the elaboration of handles, the use made of animal forms is perhaps the most notable feature. Grotesque figures are made to take the place of handles or are attached to or placed near them. The treatment is so varied that I shall have to refer the student to the subsequent series of illustrations.

Annular bases or feet were not in very general use in Chiriqui, although in some cases they are modeled with a great deal of grace. The shape varies from a simple ring, barely deep enough to give a firm support to the vessel when placed upon a level surface, to a long, attenuated column with flaring base. The latter is perhaps one of the nearest approaches which America has furnished to the slender foot characteristic of the wheel made ware of Mediterranean countries.

[Illustration: _a, b, c, d_ Fig. 52. Types of annular bases or feet--various groups of ware.]

[Illustration: _a_ Biscuit ware. _b_ Biscuit ware. _c_ Tripod group. _d_ Red line group. Fig. 53. Forms of legs--various groups of ware.]

The vessel shown in Fig. 52, _a_, has a somewhat rudimentary foot; another, _b_, a firm, wide base, which is perforated to give lightness; an hourglass-like piece is shown in _c_, and a long, bell shaped foot is seen in _d_. In no part of the world do earthen vessels exhibit such a remarkable development of legs as in Southern Central America. The tripod is the favorite support, and in Chiriqui the forms are more graceful than in the neighboring provinces. In a few cases, where the body was modeled in close imitation of animal forms, four legs were used, but three were generally preferred, even for vessels of rectangular or irregular shapes. In the simplest form they are small conical knobs, placed rather close together about the base of the vessel (Fig. 53, _a_), but from these the dimensions increase until the size is out of all reasonable proportion. The maximum development in point of expansion is seen in _b_ and the greatest height in _c_. They are frequently modeled after life forms. In a few cases rings or loops are employed, as shown in _d_. The larger forms, and especially those imitating animals, are hollow and contain round pellets of clay that rattle when the vessel is moved. The manner in which the legs are attached to the body of the vessel leads me to observe that the vessel is independently a perfect utensil, and that in all probability the tripod was a feature acquired late in the progress of Chiriquian culture, as a result of some change in the surroundings of the people or in the uses to which the vessel was devoted. Annular bases and tripods would be of little use until level floors of unyielding material came into vogue.

_Decoration._--In decoration the pottery of this province exhibits many remarkable features. The work resembles somewhat closely, in a number of its features, that of certain districts lying to the north and to the south, but at the same time it is possessed of very decided individuality. From an examination of the designs I conclude that they represent a period of culture considerably inferior to that of some more northern sections, although the ware itself is nowhere surpassed in grace of form and delicacy of finish.

The ornamentation is pretty evenly divided between plastic and flat forms. The former include relieved features and intaglio features, which are executed in the plastic clay, and the latter comprise figures in color, penciled or painted upon the surface. Each style of work embodies its own peculiar class of conceptions. Relief work is generally realistic or grotesque; incised work is almost exclusively geometric, and embraces combinations of lines usually recognized as archaic. An occasional example is easily recognized as imitative. Painted figures are both geometric and imitative, the two forms blending imperceptibly.

The more important plastic decorations consist of animal forms modeled in the round. Vegetable forms have not been employed. Fillets of clay imitating twisted cords are sparingly used in the decoration of necks and handles, and rows and groups of small nodes are similarly employed. The human figure is always treated in a conventional and usually in a grotesque manner. The animals imitated include a very large number of species. Crocodiles, pumas, armadillos, monkeys, crabs, lizards, scorpions, frogs, and fish appear very frequently. Many of the animals, owing to conventional treatment or to carelessness on the part of the modeler, are difficult of identification. These plastic forms occur in nearly all the groups of ware, and similar forms are found to a limited extent in gold, copper, and stone, as will be seen by reference to the illustrations already given. Their study will, I believe, give some insight into the mental characteristics of the Chiriquians. That their art, so far as these figures are concerned, was not serious is indicated by the sketchy, unsystematic nature of the work, and more especially by the grotesque and occasionally amusing representation of men and animals.

The figures are usually placed upon the shoulder of the vessel or are attached to the legs and handles or form part of them. The favorite subjects are doleful little figures, human or partly so, fixed upon the vessel in a sitting posture, with legs and arms doubled up, and with expressions which appear to indicate a variety of exaggerated emotions (Figs. 54, 55, 56).

[Illustration: Fig. 54. Fig. 55. Fig. 56. Grotesque figures forming the handles of small vases--terra cotta group.]

[Illustration: Fig. 57. Fig. 58. Monstrous figures, with serpent-shaped extremities--handled group.]

[Illustration: Fig. 59. Fig. 60. Fig. 61. Grotesque figures--terra cotta group.]

[Illustration: Fig. 62. Figure of monkey--terra cotta group]

[Illustration: Fig. 63. Fig. 64. Figures of monkeys--terra cotta group.]

The exuberance of fancy often found vent in the production of monstrosities, such as are seen in Figs. 57 and 58, in which the arms and legs of the figures are writhing serpents, the faces expressing great agony; in other cases the figures are double; and again two bodies united at the waist have but one pair of legs. An unusually grotesque creature is seen in Figs. 59 and 60, and another is given in Fig. 61. Similar figures are worked in gold, one of which is now worn as a charm by Mr. J. B. Stearns. Figures of monkeys are shown in Figs. 62, 63, and 64. One creature, represented as having a long, trunk-like snout, recurs frequently. Such a form discovered in the earlier days of archæologic investigation would probably have given rise to many surmises as to the contemporaneous existence of man and the elephant in Chiriqui. In reality the original was probably some unassuming little inhabitant of the isthmian jungles. This creature is shown in profile in Fig. 65, _a_, and front views are given in _b_ and _c_. Innumerable examples, embracing most of the more important animals of Chiriqui, could be given, but in a majority of cases identification is difficult or impossible, as there has been little or no effort to reproduce nature with fidelity. But the chief interest surrounding these figures is not found in the variety of creatures shown or in the character of the delineation, but in the manner of their employment in the embellishment of ceramic forms. The ancient potter must have possessed a keen sense of grace of form and of the proper adjustment of parts. The most cultured taste could hardly improve upon the lines of the vases presented in Figs. 66 and 67, which employ the frog, and in Figs. 68 and 69, in which other creatures are used. Many equally pleasing examples are illustrated further on. The question very naturally arises as to whether these little figures had any meaning or performed any function aside from that of simple decoration. I feel inclined to take the view that in their present condition they are survivals of ideographic originals; that if their past could be unveiled we would find that in the primitive ages they were not exclusively employed for ornament. The animals made use of originally were the embodiment of mythologic conceptions, and their images were revered or served as fetiches or charms, and because of this they came to have a permanent place in art. They were applied to the vessel because its office had reference to them or because they were thought to have a beneficial effect upon its functions. It is evident that their employment was governed by well established rules and that they occupied places and occurred in numbers and relations not wholly dependent upon the judgment of the individual potter. We may suppose that they occur in twos because the handles with which they were associated occurred in twos; or, if they serve to take the place of the extremities of the animal forms in the semblance of which the vases were originally modeled, their positions may be related to the original positions of the heads and tails of those forms. It is not improbable that the conventional incised and relieved ornaments, the meanders, nodes, and varied marks refer also to the creatures or the markings of the creatures with which the vessel was associated.

[Illustration: Fig. 65. Animal forms exhibiting a long proboscis--handled group.]

[Illustration: Fig. 66. Fig. 67. Vases illustrating ornamental use of animal figures--terra cotta group.]

[Illustration: Fig. 68. Fig. 69. Vases illustrating ornamental use of animal figures--terra cotta group.]

It will be seen, from the above remarks, that we cannot fully determine to what extent these ancient decorators followed the traditional pathways of early ideographic usage or how much they were governed by those powers of esthetic discrimination known to us as taste.

Unpainted Ware.

For convenience of description I separate the pottery of Chiriqui into two grand divisions: the _unpainted_ ware and the _painted_ ware. Two important groups come under the first head. The first of these, the terra cotta or biscuit ware, comprises a larger number of pieces than any other group and is readily distinguished by its colors, which include only the pale grayish yellow and reddish tints of the burned clay. The second is limited to a small number of pieces and is black or very dark upon the surface and dark within the mass.

_The terra cotta group._--This biscuit-like pottery is not in any way inferior to the painted varieties. It bears evidence of great freedom in handling, and serves, perhaps better than any other class of products, to illustrate the masterly skill and the refined taste of the ancient potter. It is said to occur in the same cemeteries and in the same graves with the more important varieties of painted ware. The function of these handsome vessels cannot be determined. It can hardly have been of a domestic nature, as they show no evidences of discoloration or wear, and we are left to speculate upon the possibility of a purely ceremonial use. The paste is moderately fine, but contains an extremely large share of gritty sand; the slip is thin and has received but a slight degree of polish, so that the surface has a dead, somewhat granular effect. As a rule the vases are of small size and are very thin walled. The forms are symmetrical and exceptionally graceful. The ornamentation includes incised figures (mostly geometric), raised decoration (of similar character), and animal forms in the round. The following illustrations are intended to epitomize the multitude of forms, as anything like a complete representation is out of the question.

[Illustration: Fig. 70. Series of bowls and cups of unpainted ware.]

Bowls, which form a leading feature of the pottery of most primitive peoples, are here rarely seen, excepting as mounted upon tripods or annular bases. There are in the collection a number of small cups of hemispherical shape that may have served as spoons, ladles, or drinking vessels. A few of these are outlined in Fig. 70. Two have minute projections resembling handles affixed to the rim. In rare cases these are so prolonged as to be of service in handling the cup; but in no instance is there an approach to the long cylindrical handles seen in the earthenware of other districts.

[Illustration: Fig. 71. Vase of graceful form--½.]

[Illustration: Fig. 72. Vase of graceful form--½.]

[Illustration: Fig. 73. Vase of fine form, ornamented with grotesque heads--½.]

In following the form scale upward from these simple shapes we find the orifice becoming more constricted and the neck more pronounced. The margins are upright, incurved, or flaring, and give variety and grace to the outlines. A tendency toward elaboration of ornament accompanies the development of form. Bands of incised or relieved figures are carried around the neck, shoulder, and handles and are added in such a way as greatly to enhance the beauty of the vessel. The forms of these vessels are so graceful and the finish is so perfect that one is tempted to present an extended series, but it will be necessary to confine the illustrations to a limited number of type specimens. Fig. 71 shows a somewhat shallow form of great simplicity and grace. That in Fig. 72 is deeper, with a narrow neck and a more decidedly conical shape. Two minute grotesque figures are perched upon the shoulder. Fig. 73 represents a larger vessel of good form, which has a neat incised pattern encircling the slightly incurved neck. Grotesque heads are set upon the shoulder. A form somewhat more refined is shown in Fig. 74. The neck is furnished with a relieved ornament, consisting of a meandered and indented fillet, accompanied by two rows of minute indented nodes. The heads are probably intended to represent the armadillo. They are hollow and contain movable pellets. The fillet ornaments are always tastefully treated, and in many cases represent twisted and plaited cords. Some are marked in herring bone fashion and others have transverse indentations. Small pellets of clay were much used and to excellent advantage. They were set on lightly with the fingers and firmly pressed down with minute pointed or edged tools and hollow straws or reeds (Figs. 75 and 76). Some of these nodes are finished to represent the heads of animals. This is done with an ease and a simplicity that call forth our admiration (Fig. 77).

[Illustration: Fig. 74. Vase of fine form, ornamented with grotesque heads--½.]

[Illustration: Fig. 75. Vase with ornaments of applied nodes and fillets--½.]

[Illustration: Fig. 76. Vase with mantle covered with incised figures--½.]

[Illustration: Fig. 77. Vase with frieze of grotesque heads--½.]

Fig. 78 illustrates a series of vases having flaring rims, the treatment otherwise being uniform with the preceding. We notice in these vessels a decided tendency towards complexity of outline. Three examples, shown in Fig. 79, have a two storied character, the upper part possibly being the outgrowth of the collar ornament seen in so many cases. The large specimen in the center is a handsome piece with square offset at the shoulder and a decidedly conical base. A chaste ornament in relief encircles the neck and two grotesque figures are seated upon opposite sides of the shoulder. The vase at the left has two orifices, set wide apart. The body is oblong and slightly flattened above. There are a number of vessels of this conformation in the collection, some of which have the mouths so close together that the margins or lips coalesce in part. A superb specimen of this class is illustrated in Fig. 80. The shape is thoroughly satisfactory to the eye, having a refinement of line rarely attained in native American work. Its symmetry suggests the use of the wheel, but the closest examination fails to detect a trace of mechanical appliance, save that left by the polishing stone. The decoration is simple and effective, consisting of minute nodes with annular indentations about the necks and of two grotesque figures, placed with consummate taste in the angles formed by the contact of the two necks.

[Illustration: Fig. 78. Vases with flaring rims and varied ornament--1/5.]

[Illustration: Fig. 79. Vases with complex outlines and varied ornament--1/3.]

[Illustration: Fig. 80. Large vase with two mouths and neatly decorated necks--½.]

[Illustration: Fig. 81. Large vase with high handles--1/3.]

A very small percentage of these vessels possess true handles, but these, in some of the examples, are worthy of high admiration. The specimen presented in Fig. 81 attracts attention at once on account of its resemblance to well known classic forms. It is evident, from a study of this piece, that only a step more was necessary to place these potters alongside of the highest masters of the art. The sharp high elbow and the broadening of the handles at their junction with the lip are notable features. The latter is shown more satisfactorily in Fig. 82, which is a top view of a companion piece. I wish to call attention here to a peculiar feature of these handles and one repeated in vessels of other classes. At the elbow of each handle we find a device in relief marked with herring bone indentations that would seem to represent a kind of textile attachment, as if, at some previous time and perhaps in an antecedent form of vessel, the upright and horizontal parts of the handles had been stitched or tied together at this point. Yet it is by no means certain that this feature is not the survival of some feature of an animal form into the semblance of which, as seen in other examples, this feature has a tendency to graduate.

[Illustration: Fig. 82. Top view of high handled vase--1/3.]

[Illustration: Fig. 83. Fig. 84. Fig. 85. Examples of handled vases--½.]

These vessels are not numerous, but acquire importance from their large size, the larger being upwards of eight inches in height. A few pieces of nearly identical shape, but of small size, are found among the painted wares. Additional shapes are given in Figs. 83, 84, and 85, and serve to illustrate the extent of variation exhibited in this group of vases. The small shallow piece is exceptionally fine and the handles are furnished with animal features of a highly conventional type. An expansion of the handles somewhat similar to this is frequently seen in vessels of other classes, especially in those of the handled group.

[Illustration: Fig. 86. Fig. 87. Small cup with single handle ornamented with grotesque figure--½.]

Single handles of like character occur in a number of cases upon minute cups, an example of which is given in Fig. 86. It would seem that possibly in such cases the rim had been expanded and prolonged for the purpose of giving support to the animal figures with which the shoulders were embellished. The expansion is probably the outgrowth of the use of animal figures in connection with simple handles.

[Illustration: Fig. 88. Vase of eccentric form--½.]

We have a number of vessels of this group the bodies of which imitate animal forms, but they are in nearly all cases furnished with legs. Rarely we meet with compound or eccentric forms. An interesting specimen of the latter class is seen in Fig. 88. Such shapes are common in Peru and are occasionally met with in Central America. The two strong handles are decorated with minute images of birds and the bottom is concave, an exceptional character in Chiriquian work.

The illustration of this group of vessels would not be complete without a series of tripod vases. In shape of body these vases differ but little from the legless forms already given, excepting where the use of life forms has led to eccentric modifications. Very great interest attaches to the modeling of the tripod supports, upon which the potters have expended much time and ingenuity.

[Illustration: Fig. 89. Fig. 90. Vessels illustrating forms of legs--½.]

[Illustration: Fig. 91. Vessel with large legs ornamented with stellar punctures--½.]

[Illustration: Fig. 92. Vases of varied form with plain and with animal shaped legs--1/3.]

The illustrations given herewith are chosen from a great number of examples and are intended to convey an idea of the range of forms, both of the vessels and of their supports. Figs. 89 and 90 show plain forms of legs, all of which are hollow and contain small pellets of clay. The openings are generally wide vertical slits, and are placed in front, as seen in Fig. 89, or in the side, as in Fig. 90; but in exceptional cases they take other shapes and are scattered over the surface, as seen in Fig. 91. The legs are often remarkable in form, being swollen to an enormous size above and terminating in small rounded points below. The bowls are symmetrically shaped and graceful in outline. In Fig. 92 I present a group illustrating some of the more eccentric forms of bowls and a variety of their supports. A very superior piece and one of the largest of this style is shown in Fig. 93.

[Illustration: Fig. 93. Large vase of striking shape--½.]

[Illustration: Fig. 94. Cup with legs imitating animal forms--½.]

It will be seen that in a number of cases the legs are modeled to represent animal forms. This feature is brought out more clearly in succeeding figures. The creatures represented are often grotesque, as seen in Figs. 94 and 95. The human form is rarely shown in a way to make it clearly distinguishable from the figures of monkeys and other animals. The armadillo is a favorite subject. An example of small dimensions is illustrated in Fig. 96, in which this animal is given in characteristic style, and a more pretentious piece is shown in Fig. 97. The characteristics of the creature are very simply but graphically expressed. In the first the hard ribbed and figured case is represented by applied fillets and nodes, and in the other by incised lines. The frog is also much used (Fig. 98). A rather remarkable conception is illustrated in Fig. 99. Upon the front of each leg is a curious little animal-like figure, to the front of which are bound two minute infantile creatures. In the piece presented in Fig. 100, the legs are grotesque heads, inverted, with wide open mouths and glaring eyes. The work upon this vase is very superior.

[Illustration: Fig. 95. Cup with legs imitating a grotesque animal form--½.]

[Illustration: Fig. 96. Fig. 97. Cups with legs imitating the armadillo--½.]

[Illustration: Fig. 98. Cup with frog shaped legs--½.]

[Illustration: Fig. 99. Cup with legs imitating an animal and its young--½.]

[Illustration: Fig. 100. Cup supported by grotesque heads--½.]

The remarkable specimen illustrated in Fig. 101 is furnished with unique supports. Two rudely modeled, semihuman, grotesque figures are affixed to the under surface of the bowl, supporting it with their backs.

[Illustration: Fig. 101. Large cup supported by two grotesque figures--1/3.]

[Illustration: Fig. 102. Cup with two animal heads attached to the sides--¼.]

[Illustration: Fig. 103. Cup with two animal heads attached to the sides--¼.]

The legs of these figures are spread out horizontally, so that a firm support is obtained. The periphery of the body of this vessel is encircled by a number of nodes and noded projections, which represent the heads, tails, and spines of two crab-like animals. The heads, with arms attached, appear at the right and left, and the tails occur at the front and back just over the heads of the supporting figures. The use of the crab in this way is quite common. Fish, birds, and a variety of quadrupeds are similarly treated. Some very interesting examples of double headed animal vases are found. Two of these are outlined in Figs. 102 and 103, the first having a single orifice and the second a pair of orifices. In many cases the bowl of