ADDITIONS AND ORNAMENT. 137 
possibly an evolution upon the cupped depressions so frequent on the 
missile clubs; it differs therefrom in having a flat floor and sides vertical 
thereto. Between the limiting circles we find that the artist has dealt 
with great vigor in the problem of ornament. In the radiant lines he 
varies but slightly from the trisection of the circle; the left radiant is 
exactly 120° from the upper; the right is within 5° of mathematical 
accuracy, a remarkable feat for a man working solely by the eye and 
without dividers, with no knowledge of the constant 7, ev-n ignorant 
of the use of a piece of cord for measurement. Not only are these 
points established with satisfactory precision, but in the treatment of 
the three wedge-shaped bodies he displays a recognition of the diver- 
gence of the radii of the circle. There is commendable vivacity in his 
treatment of the concentric arcs by setting them in panels and in his 
finish of the whole composition by the addition of an outer circle com- 
plete. In figure 49 much cruder work appears, two concentric rings of 
band-and-zigzag, with the suggestion of an inner unit of the same 
berieath the obscuring four-rayed figure, the curves being most uncer- 
tain and the angles of the center piece quite away from the rectangular 
crossing. 
The designer of the latter piece gives two more circles on the face of 
the shaft; one (fig. 47) a plain ring with 5 radii irregularly spaced, the 
other (fig. 48) with 4 equally irregularly spaced radii and a concentric 
ring of poor zigzag and an outer plain ring. To another designer, in 
_ figures 46 and 47, are attributed two similar figures of double concen- 
tric circles about a central circular spot, which in the latter is consider- 
ably distorted, both of these figures partaking of the general coarseness 
of design upon this piece. In figure 43, still by the same artist, an 
instance occurs of his avoidance of a similar circle of triple concentric 
lines by a diametrical erasure such as is seen in the lozenges of figure 26; 
in this case it is plain that the figure derives from the opposition of two 
such figures of concentric angles as found at the top of the panel, and 
the line of erasure is intended to hold the design to that standard. In 
the decoration of the great serrated club in Plate I f is found the em- 
ployment of plain disks arranged quarterly on each face of the head; 
that at the upper left of the illustration is distinguished by an inner 
concentric circle quite near the edge. 
On four clubs four instances of an arcuate figure occur, two single 
and convex outward, two convex inward and double and triple respec- 
tively. If there were no more than the single outward arcs in figures 
50 and 51 we might regard them as sky symbols. So little intensive 
study has been directed upon art motives among the Pacific islanders 
that we are unable to establish the employment of such a figure as por- 
traying thesky. Our utmost information as to the heavens comes from 
literature rather than from art; we have abundant proof in the myths 
that the sky is looked upon as an inverted bowl. From almost every 
