354 DR W. CARMICHAEL M'INTOSH ON THE STRUCTURE OF THE 



ledo-e of its relations is obtained. I have not been able to see 0. alba feed 

 in captivity, and have not found any food in the alimentary cavities of those 

 examined. It is a curious fact, that in this group the digestive system lies quite 

 beneath the nervous system anteriorly, whereas the proboscis passes through the 

 nervous collar. 



Circulatory System. — The circulatory system is composed of three great longi- 

 tudinal trunks — one central and two lateral— besides the cephalic arch and 

 anastomotic vessels. Commencing with the great central trunk posteriorly 

 (Plate VI. fig. S,p) in Ommatoplea, it is found that the vessel, which in this region 

 is about twice the diameter of the lateral, arises from the point of junction of the 

 two last-mentioned, just within the posterior border of the worm. It travels 

 forward beneath the proboscidian chamber in a very undulated manner— as 

 usually seen— to the region behind the ganglionic commissures, where it bifurcates 

 (Plate VI. fig. 3, q), a branch passing to either side to join the lateral trunk (r), 

 which bends inwards to meet it. From this point of junction also a single 

 vascular arch (cephalic) proceeds forwards into the tissues of the snout (/, same 

 figure, and in Plate IV. fig. 6, the latter showing the vessels in transverse section), 

 the pillars of the arch thus meeting the lateral and the anastomotic vessels of 

 each side. From the same point of union each lateral trunk passes backwards 

 under the nerve-cord of its side to the tail, where it meets its fellow of the opposite 

 side, and gives origin to the single central vessel with which the circuit com- 

 menced. The lateral vessels appear to diminish slightly posteriorily. The 

 median vessel does not actually touch the wall of the proboscidian sheath, though 

 transverse sections usually show a close apposition, but is situated in a layer of 

 transparent elastic tissue which intervenes between this organ and the digestive 

 tract. At the ganglionic region the vessels which go to form the cephalic arch 

 pass below the commissures, and unite in front beneath the channel of the 

 snout. In 0. purpurea there are three main longitudinal trunks as in 0. alba; 

 but it can be observed that the lateral communicate with the central, as in Borlasia. 

 by transverse branches, which, however, are proportionally smaller. Whether 

 such anastomoses occur in the pale Ommatopleans is thus an open question; 

 but they are distinct enough in this species. Two lateral trunks only could 

 be discovered in Polia involuta, Van Beneden (Plate VIII. fig. 5, r), which 

 trunks unite by a very short loop just in front of the commissures. This loop (/) 

 is distinguished from the ordinary arrangement by its not extending forwards 

 into the tissues of the snout. The lateral vessels are not so clear or well defined 

 as in 0. alba and Tetrastemma, and are observed to have internal transverse 

 bands or partial septa in front ; while the contained fluid has a few clear granules, 

 as in 0. purpurea and others. The contractions in the lateral vessels are very 

 vigorous, and even a minute central vessel could not have been passed over if a 

 trace of such had existed. 



