THE METAMORPHOSIS OF ECHINODERMS. 51 
Third Ciliated Ring.—This is formed as simply as any, 
from one piece on each side (occasionally two on the left side, 
the ventral one being small, and soon joining the dorsal one). 
In stage B (fig. 5) they join on the ventral side, but remain 
apart much longer on the dorsal side, where they eventually 
unite in front of the water-pore. 
First and Second Ciliated Rings.—Figs. 3 and 6 will 
show that we have now four pieces of the ciliated band with 
which to construct the two anterior rings. At first they form 
two calliper-like loops, one on each side (fig. 6), and this 
arrangement is still partly retained in stage A, of which fig. 7 
is a polar view. 
In stage B (figs. 5 and 8) the opening of the atrium has 
moved into the left loop, and it seems as if this loop were 
about to form the first ciliated ring. Fig. 9, however, shows 
that only the ventral portion of the loop is concerned in form- 
ing this ring, while three pieces (II a, 6, and ¢ in figs. 7, 
8, and 9) form the second ring. The two pieces of the right 
side (II 6 and c) usually unite first (occasionally II ¢ is 
absent), and a little later the right and left sides unite 
dorsally ; lastly they unite on the ventral side, and the ring 
is complete. 
The investigation of the phenomena just described requires 
a good deal of care and patience. The rapidity of the changes 
is one of the most serious difficulties, necessitating a rigorous 
search each day among one’s specimens for signs of the im- 
pending change. At the time when most of my observations 
were made (Naples, March and April, 1888) the stage shown 
in figs. 2 and 3 was usually reached about ten o’clock on one 
morning, stage A late that evening, and the stage shown in 
fig. 9 about ten o’clock the next morning; after that the 
changes were slower. Development was much slowed down 
by keeping the larva in cold water, and in this way I was 
able to obtain all the intermediate stages without resorting 
to twenty-four hours’ consecutive watching. 
Another difficulty lies in the shape of the larva, which 
makes it hard to obtain prolonged views of any but the dorsal 
