118 CLUB TYPES OF NUCLEAR POLYNESIA. 
face of the head and disregards the suggestion present in the other 
species of a shaft socket against which the blade is fixed. The rugose 
panel consists of two members—transverse wrinkling across the whole 
of the inner curve and less conspicuous scoring, the general effect being 
such as might be produced in less dense timber by exposure in a steam- 
box and bending under great pressure. ‘The rugosity is so clearly 
diagnostic that we shall have no difficulty in recognizing its probable 
source. The outer surface of the tridacna is heavily sculptured in 
just such wrinkles, accompanied by less conspicuous lines generally at 
right angles thereto, the wrinkles of the shell being concentric with the 
edges of the lip. Blades occur quite frequently with the natural sur- 
face on one side. This must rest upon observation, for the many 
specimens in the museum are highly polished upon both faces and do 
not present this character. There is no part of the shell, not even at 
the very dense and on that account preferable hinge region, which can 
exhibit a rugosity such as in this species of club continues from one 
face over the other. It appears to me that if there were no more than 
the single face such as is seen in figure c, we have a satisfactory inter- 
pretation of the motive in a slip of shell with its natural surface and 
ground down at the end to a cutting-blade inserted on a shaft carved 
to receive it; the rugosity on the other face will then be comprehended 
as motivated by the general principle of symmetry in design. So 
many of the critical units of the type are lacking to this species that we 
may regard it as secondary evolution, but the persistence of the blade 
is quite sufficient to warrant its inclusion in the type. It will be appar- 
ent-in the illustrations, far more conspicuously manifest in the pieces 
themselves, that in this species we have passed away from the lightness 
of the two foregoing species, which suggested greater importance of the 
cutting-edge of the blade, and have here produced a weapon whose 
impact force is reached by added weight grouped in the head. 
In all three species of this type the blade continues on the face of the 
head from its lower point up to the middle of the upper edge. This is 
clearly a convention. In the two earlier species, in which the dis- 
inction between blade material and shaft-socket is structurally indi- 
cated, it will be apparent that a really structural cutting-edge must 
stop short at the beginning of the shaft-socket. But in the extension 
of this unit from utility to decoration we find no difficulty in the exten- 
sion of the edge beyond its structural possibility. 
The axe-bit club is of a type of extreme complexity (Plate II, fig. e; 
Plate IV, figs. 1-3). Considered as a whole, it presents the general 
appearance of a curved club; yet that can be proved more apparent 
than real. I dissect the piece into three units—shaft, socket, and 
blade. 
The shaft is simple, a rectilinear column of oval section ending dis- 
tally in a strongly marked shoulder set diagonally to the length and 
