26 
THE VOYAGE OE H.M.S. CHALLENGER. 
simple, long, slender, and finely tapered bristles wliicli have tlie tips (PL IIIa, fig. l) 
minutely roughened, from the point a considerable way downward. Such microscopic 
processes on the surface are homologous with the bolder condition in Amphinome. The 
other kind of bristle is distinctly serrated at the tip (PI. IIIa. fig. 2), upwards of twenty 
teeth being visible, and a fold of the chitine passes from the apex of each serrature down- 
ward like a hood. These bristles show in a remarkable degree the albuminous contents 
(the brownish masses of Grube) after drying, for each bristle collapses, and the contents 
form nodular masses all over the exterior. Glacial acetic acid causes many bubbles of gas 
to escape from the central cavity of the bristles, but has no effect on the globules. 
Sulphuric ether and absolute alcohol show as little action on the latter as the acid. 
They would therefore appear to be alb umin ous, though their appearance is decidedly 
fatty. 
The ventral bristles are all of one kind (PL IIIa. fig. 4, which represents one of the 
inferior bristles), presenting a slightly striated shaft and a curved, slightly bifid tip, 
while a series of very distinct serrations occur on the edge of the latter, from the tip 
to the spur. Like the dorsal bristles, each of the points or serrations does not indicate 
merely a spike, but also a slight wing. The upper bristles possess much longer and 
more slender tips, and the spike is more distinct (PL IIIa. fig. 3). In the bleached 
example from Bermuda the serrations and tips of the ventral bristles are much abraded, 
so that only those sheltered by position show the ordinary structure. The animal, 
therefore, makes some use of them, for progression or otherwise, and probably under 
stones. In the specimens from St. Vincent, Cape Verde Islands, these bristles form a 
somewhat regular vertical row, the longest bristles (with the longest tips) being 
superior. In the very large example many of the ventral bristles are more distinctly 
bifid than usual. 
The dorsal cirrus arises behind the bristle-23apilla, and has a very attenuated tip. It is 
composed of a basal and a distal division. The ventral cirrus is smaller and occupies the 
ordinary position behind the ventral bristle-tuft. Both increase in length posteriorly. 
In the stomach of the large sj)ec im en were muddy masses containing sponge-spicules, 
Diatoms, Eadiolarians, sand-grains, fragments of chitine, and cellular and granular 
debris. 
The nerve-cords are quite within the body-wall, for the oblique muscles meet in the 
middle line beneath them, the fibres apparently decussating vdth the circular muscular 
coat. Connective tissue and granular matter, moreover, separate them from the muscular 
fibres, and in the sections the perivisceral corpuscles also often intervene. A considerable 
hy|3odermic canal occurs immediately below the decussation just mentioned. This seems 
to be filled with opaque material. When a section is made in the line of the ganglia, a 
large nerve-cord on each side is seen to arise from the cells in their interior. The upper 
and lower walls are strengthened by a dense capsule. Superiorly also a peculiar conical 
