Continuing in his "Ancient Art in Chiriqui,"[300] presenting his "Series showing stages in the simplification of animal characters," and "derivation of the alligator," Professor Holmes elaborates the theory how the alligator was the original, and out of it, by evolution, grew the cross. His language and accompanying figures are quoted: Of all the animal forms utilized by the Chiriquians, the alligator is the best suited to the purpose of this study, as it is presented most frequently and in the most varied forms. In figs. 257 and 258 [figs. 330 and 331 in the present paper] I reproduce drawings from the outer surface of a tripod bowl of the lost color group. Simple and formal as these figures are, the characteristic features of the creature--the sinuous body, the strong jaws, the upturned snout, the feet, and the scales--are forcibly expressed. It is not to be assumed that these examples represent the best delineative skill of the Chiriquian artist. The native painter must have executed very much superior work upon the more usual delineating surfaces, such as bark and skins. The examples here shown have already experienced decided changes through the constraints of the ceramic art, but are the most graphic delineations preserved to us. They are free-hand products, executed by mere decorators, perhaps by women, who were servile copyists of the forms employed by those skilled in sacred art. A third illustration from the same group of ware, given in fig. 259 [fig. 332 of the present paper] shows, in some respects, a higher degree of convention. * * * I shall now call attention to some important individualized or well-defined agencies of convention. First, and most potent, may be mentioned the enforced limits of the spaces to be decorated, which spaces take shape independently of the subject to be inserted. When the figures must occupy a narrow zone, they are elongated; when they must occupy a square, they are restricted longitudinally, and when they occupy a circle, they are of necessity coiled up. Fig. 265 [fig. 333 of the present paper] illustrates the effect produced by crowding the oblong figure into a short rectangular space. The head is turned back over the body and the tail is thrown down along the side of the space. In fig. 266 [fig. 334 of the present paper] the figure occupies a circle and is, in consequence, closely coiled up, giving the effect of a serpent rather than an alligator. * * * [Illustration: Fig. 332. CONVENTIONAL FIGURE OF ALLIGATOR. From a vessel of the lost color group. Chiriqui. Sixth Annual Report of the Bureau of Ethnology, fig. 259.] [Illustration: Fig. 333. CONVENTIONAL FIGURE OF ALLIGATOR CROWDED INTO A SMALL GEOMETRICAL FIGURE. Chiriqui. Sixth Annual Report of the Bureau of Ethnology, fig. 265.]