HARTMAN: ARCHEOLOGICAL RESEARCHES ON THE PACIFIC COAST OF COSTA RICA 71 
vex, while the convex upper half of the front shows in high relief the head and 
wings of the bird, and the lower half is cut flat and thin and is devoid of ornamen- 
tation. (Compare the figures on Pl. XXXVI.) The latter type in many specimens 
takes the form of a grooved ax, the head of the bird serving as the head of the ax, 
the neck representing the grooved part, and the lower half of the body forming the 
blade, which often broadens out toward the edge. In the more highly convention- 
alized forms of similar celts the bird-like features gradually disappear. In a num- 
ber of these small amulets all avian features have been dropped, and only a plain 
grooved ax is left. Pl. XXXVI shows a number of the transitional forms. ‘The 
crests and ear-tufts of the bird are generally emphasized by the artist. An incised 
band at the front of the head, similar to the head-band of the human figure, is fre- 
quently seen, and some birds are adorned with high head-dresses resembling those 
on the anthropomorphic celts. In some figures, as those represented on Pl. XX XV, 
Figs. 19, 21, 22, and 25, the feathers of the wings are delineated, and a few specimens, 
as those on Pl. XX XV, Figs. 21 and 25, even show the legs and the feet. As most 
of the bird-celts are more or less convex and rather thick, the hole 
for suspension is nearly always drilled from the side ; only in a few 
very thin specimens which could not be pierced in this way, are two 
circular holes drilled in from the front. 
In a certain portion of the ornithomorphic celts (only rarely 
in the anthropomorphic), the figure, instead of appearing as 
seen ‘en face’ is sculptured in profile, with the head cut out 
usually on the left, but in some cases on the right side. On PI. 
XXXVII are shown a number of specimens of the latter class, 
ranging from examples in which all the details of the body are 
brought out to such as represent the most conventionalized and sim- 
plified forms. Two similar figures are also shown on Pl. XLIV, 
Figs. 2 and 4. Fie. 70. Bird- 
The zodlogical identification of conventionalized animal forms is Shaped celt, wrongly 
identified as a whale 
often a puzzling task, particularly when the investigator does not by Dr. H. Fischer. 
have at his disposition large series of all the animal forms repre- 
sented in art by the people or culture under study. Even then it frequently is 
impossible in certain cases to make exact determinations, owing to the fact that in 
art different animals have been given purely fantastic characters, and hybrid forms 
of animals, or of man with animals occur. ‘To draw conclusions from a small num- 
ber of specimens in consequence becomes a risky undertaking. <A wide field for 
speculation is always open to the student who tries such an experiment. In Dr. 
