HYALONEMA (PRIONEMA) FIMBRIATUM. 303 
curved, usually all more or less in the same direction. In short spines this 
curvature is insignificant, in long spines, very pronounced. The long spines 
are usually strongly curved in their basal part and straight or slightly curved, 
not infrequently in the opposite direction, in their distal part. The plane of the 
main curvature either passes through the axis of the shaft, or it is oblique to it. 
Since, as above stated, all the spines on the tyle are usually curved in the same 
direction, the verticillate bunch formed by them appears — when the spines are 
long — bent, straight or spirally, toward one end of the spicule. The remain- 
ing parts of the shaft are smooth. The axial thread passes through the central 
tyle without being thickened, and there is no trace of an axial cross. I noted, 
however, a few small dots in the central part of the tyle, near the axial thread, 
which appeared to have the same refractive index as the axial thread. 
The two anchors of the same spicule are fairly equal, or rather unequal, in 
size. The anchors are 80-180 u» long, a third to two fifths of the whole spicule, 
and 70-113 4 broad. The proportion of anchor-length to anchor-breadth is 
100 to 46-84, usually 100 to 63-80, on an average (of sixteen calculated individual 
proportions) 100 : 70. 
The individual anchor-teeth are near the base 10-14 u high, and a little 
farther out, at their widest point, 12-22 4 broad. They are attenuated distally 
and at the end simply rounded (Plate 59, figs. 5, 6; Plate 62, fig. 31) or, more 
rarely, divided, by a slight indenture, into two terminal lobes (Plate 59, fig. 4). 
They arise steeply or vertically from the end of the shaft and are curved in their 
basal part through an angle of about 80°. Then they become nearly or quite 
straight and remain so to within a short distance of the end. This long straight 
part of the tooth ses an angle of about 10° with the continuation of the axis 
of the shaft. The end-part of the tooth is slightly bent outward or more rarely 
straight and excended in the same direction as the middle-part. The plane of 
the normal curvature of the tooth passes through the axis of the shaft. Occa- 
sionally the end-part of a tooth is bent also in a plane vertical to this, paratan- 
gentially as it were, to one side (Plate 59, fig. 6). 
The fimbriate amphidiscs are all very similar. The only differences between 
the four biometric groups of them, that is the largest, large, small, and smallest, 
are those due to the relative anchor-breadth decreasing and the relative size of 
the central tyle of the shaft increasing with the size of the spicule. Their anch- 
ors are from a little less than two fifths to nearly half the whole spicule in 
length. 
The largest fimbriate amphidiscs (Plate 60, figs. 1-6, 24; Plate 61, figs. 1-11; 
