LARVAL SKELETON OF SPATANGUS 487 
is directed anteriorly, another postero-laterally, and the 
remaining one postero-medially. From the lack of adequate 
material the fate of the former two arms cannot be stated with 
certainty, though it seems probable that they do not develop 
much farther. The postero-medially-directed arm in the later 
stages continues to develop in a direction parallel to the dorsal 
surface, reminding one of the body-rod on the ventral side 
(figs. 7 and 8). Near the base of this arm a branch is sent out 
in an antero-median direction, reminding one again of the 
ventral horizontal rod. This is the dorsal horizontal rod (dh). 
From each of the arms of the star, close to the centre, is 
given out a vertical process, very often differing in the rate of 
development, but ultimately the three in all give rise to the 
latticed postero-dorsal rod. 
Although from want of material, especially of the later 
part of this stage, no definite statement can be made, yet, 
judging from later specimens, it is highly probable that 
the post-oral rod increases in length during the six-armed 
Stage up to nearly 6 times the length of the body-rod, the 
antero-lateral rod 3 times, the aboral spike nearly 3-5 times, 
and the postero-dorsal rod probably at least 1-5-2 times the 
length of the same. 
In Krohn’s figure (2, Pl. vii, fig. 3) is indicated the three- 
rayed base of the postero-dorsal rod (e). The buds of the 
pre-oral arms have already appeared (d), while the dorsal arch 
is still in a rudimentary condition, of which, however, nothing 
is mentioned. The fact that the pre-oral arms appear without 
any mechanical influence of the underlying skeleton is also 
seenin Echinocardium cordatum (4, pear? Plo exci, 
fig. 6). But both in MacBride’s case of Echino- 
cardium and my specimens of Spatangus the appear- 
ance of the pre-oral arms takes place much later than the 
stage as shown by Krohn, viz. even when the postero- 
dorsal arms have attained a fair length, there was as yet no 
sign of these arms found. Krohn gives some detailed 
structures in a somewhat advanced six-armed stage (Pl. vii, 
figs. 5 and 6). If the fig. 5 is really the dorsal view, as stated 
