BRANCHIURA SOWERBYI BEDDARD. 291 
the dorsal gill; and one ventralwards (v.d.), with a similar course and distribution on 
the ventral side. The two branches may, in the region of the anterior gills, arise in 
common (fig. 7). The ventral vessel gives two branches: one goes to the right and 
bifurcates almost immediately into two divisions, which correspond to the two branches 
of the dorsal vessel,—one running dorsalwards on the inside of the body-wall to the 
dorsal gill (d./7.v.), the other taking a short course ventralwards to the ventral gill 
-(v.br.v.). The second branch of the ventral vessel (/.b7.v.), leaving the main trunk 
on the left side of the latter, arches over the nerve cord and goes to the left side of 
the body-wall; it gives a branch to the right below the nerve cord, which seems, 
according to sections, sometimes at least to meet and anastomose with the branch 
from the ventral vessel to the ventral gill; thus in these cases a ring is formed round 
the ventral nerve cord. 
I am not satisfied that the arrangement above described is constant; in one series 
of sections the two vessels going to the dorsal gill were derived, one from the dorsal 
vessel in the usual way, described above, and the other from the branch from the 
ventral vessel to the left side of the body-wall (7.e. the dorsally directed branch of 
the vessel /.b7.v. in fig. 8 went up into the gill). 
In front of the gills the parietal vessels have the form of lateral loops, two per 
segment,—one on the anterior face of the septum, the other more anteriorly in the 
segment. ‘There are also, as described by Bepparp, a number of longitudinal vessels, 
some of which connect successive loops, while others, smaller, pass over several 
segments at least. ‘The muscular layers of the body-wall are penetrated by numerous 
twigs, but no capillaries seem ever to enter into the surface epithelium. 
The swpra-intestinal vessel is hardly or not at all distinguishable in front of the 
sixth segment, and it is small as far back as segment viii; here two considerable loops, 
one on each side, put it in connection with the ‘anterior ventral’ vessel ; these loops, 
though not as large as the hearts in the next segment, are still extremely conspicuous 
structures. In segment ix it gives rise to the first pair of hearts, and in the same 
segment a number of vessels radiate from it across the body-cavity to the parietes. 
In this part of its extent it lies directly beneath the dorsal vessel, and, like the latter, 
gets round to the left side in the posterior part of its course. It is largest in segments 
xX and xi; it communicates by wide channels with the intestinal plexus, and is 
throughout in close relation with the gut and its chloragogen cells. It becomes 
smaller in segment xii and very soon disappears. 
The hearts are two pairs, in segments ix and x. The first pair originate above 
from the supra-intestinal vessel, pass downwards close to the gut, then, taking a 
backward course, leave the intestine and perforate septum ,% separately ; continuing 
parallel for a short distance, they then unite in x to form the ventral vessel. The 
second pair appear to be more tortuous and of rather smaller calibre; they originate 
above from the dorsal vessel, which is here on the left side, so that the heart of the 
right side arches over the intestine; they pierce septum 1+ separately (this septum 
b>. 
