292 DR J. STEPHENSON ON 
thus transmits three vessels ventral to the intestine) and unite with the ventral vessel 
in xi. Where the hearts join in the ventral vessel there are in both segments (x and 
xi) projections of cells into the lumen; these may, according to the evidence of 
sections, entirely block the cavity of the vessel. 
Lastly, the anterior loops occur in segments ii-viii; they are non-contractile, 
and the posterior are larger than the anterior; in vii they are of about the same size 
as the dorsal or the ventral vessel, and in vill they are so large as to resemble the 
hearts. In ii-vii they run from the dorsal vessel to the anterior ventral, or to the 
branches which unite to form the latter; in viii, as already said, they run from the 
supra-intestinal to the anterior ventral. They give branches to the body-wall, which 
in vi, vii, and viii are of moderately large size. It may be added that the dorsal 
vessel also gives considerable branches to the body-wall in vi and vii, and the parietal 
vessels form a plexus in the prostomium and most anterior segments. Behind segment 
iv the parietal vessels have a longitudinal course, running parallel and fairly close — 
together, so that about 20 such vessels are visible on examining the dorsal surface of 
the worm; there would thus be about 40 longitudinal vessels in all at any particular 
level, each of which runs through several segments without losing its individuality ; 
this arrangement, as before noticed, ceases at the gill region. 
The only other feature which I propose to select for special mention is the presence 
of remarkable giant fibres in the ventra] nerve cord. They may be seen throughout 
the body on the dorsal side of the cord. They are of different sizes; the larger appear 
as tubes in transverse section, with a roughly circular or oval outline; a part of 
their lumen is always empty, and a part, to one side, is occupied by a substance which 
sometimes has the appearance of a fairly solid mass, sometimes that of a thin” 
coagulum ; the shape of this contained matter varies, being sometimes ovoid, sometimes 
quite irregular; at times it is connected here and there with the wall of the tube by 
a number of branching threadlike extensions; sometimes it occupies a considerable 
portion, sometimes very little of the lumen (figs. 38, 5, 6). Other fibres, also running 
longitudinally on the dorsal and dorso-lateral surface of the cord, are not tubular, and 
in their staining reactions somewhat resemble muscle fibres. 
The number of these large fibres varies from place to place; there are in all, in any 
section, about five to ten; they are most often six or seven in number, of which one to 
three have the tubular appearance described above, the rest being larger or smaller 
solid fibres. . 
The actual size of the tubular fibres also varies. One is generally larger than the 
others,—in the posterior part of the body very much larger,—and this one is there 
constantly to the left of the middle line. These fibres are on the whole larger towards 
the hinder end of the animal; in the anterior segments they average about 20# in 
diameter; as the gill region is approached they may be 40«; while posteriorly one, but 
only one at any given level, may be as much as 71; these large fibres are, however, 
much constricted at the septa in this region, ¢.g. to about 154 or even much less, 
