Respiration in Invertebrate Animals. 403 



the edges of the respiratory laminae : these cilia are most cou- 

 spicuous in P. lamelligera. This is a sti'iking point of distinction 

 between the Phyllododdce and the Nereids, in which ciliary epi- 

 thelium has no existence. The chylaqueous tiuid then may be 

 clearly affirmed as that agent in the oeconomy of the Phyllodo- 

 cidcE which is the immediate, the first, subject of the respiratory 

 process, the true blood receiving its supply of oxygen from this 

 fluid, afterwards to convey it to the solid structures of the body. 



In the genus Glycera the blood-proper is entirely excluded 

 from the organs of respiration : this office devolves exclusively 

 on the chylaqueous fluid. The latter in this worm is crowdedly 

 charged with ?-e</-corpuscles, a remarkable exception to the An- 

 nelidan rule. The gills consist of hollow, cylindrical appendages, 

 emanating from the base of each dorsal foot at its superior aspect, 

 filled in the interior with the chylaqueous fluid. It is peculiar 

 to and distinctive of this genus that the interior of the branchise 

 is lined with vibratory epithelium. Cilia are not detectible on 

 the exterior of these processes, but they exist in the interior : 

 under the action of these oar lets, the corpuscles of the chylaqueous 

 fluid by which the gill-process is penetrated, move with great 

 rajjidity in a definite dii'ection; — peripheradly on one side and 

 centradly along the other, each corpuscle whirling on its own axis 

 as it proceeds. The advancing stream, however, is not divided 

 from the returning. The channel is one, open channel. This 

 is the law which is apphed to the chylaqueous fluid : it knows 

 no exception. Its channels are always single and csecal : its 

 movements a flux and reflux. Contrarily the orbit of the true 

 blood is circle-like, its channels closed, its colour red, its compo- 

 sition non-morphotic. 



In the Syllidce (fig. 6) the branchial organs are penetrated only 

 by the chylaqueous fluid. It can be detected only at the bases of 

 the feet {d). To this part the vibratile cilia are restricted (c). 

 The long filiform, and in some species moniliform or leafy appen- 

 dages which are described commonly as the branchise in these 

 worms, have no central hollow ; they are filled with lacunose 

 tissue (/) through which the fluid parts of the contents of the 

 visceral chamber slowly penetrate. But in the spacious cham- 

 bers occupying the bases of the feet, the corpuscles may be detected 

 in whirlpools. From this fact the inference may be drawn that 

 the corpuscles are not essential to the first mechanical stage in the 

 respiratory process — that of receiving oxygen directly from with- 

 out. The characters of structure just described are very perfectly 

 typified in S. prolifera, the moniliform variety is best seen in 

 S. armillaris and S. maculosa. A similar conformation pre- 

 vails in the genera loida and Psamathe of Dr. Johnston. In 

 the Syllidau family, which excels all others in grace and beauty, 



28* 



