REPRODUCTION OF LOST PARTS. 125 



the long anterior tuft of cilia by the young Cephahthriae has its analogue in the loss of the ciliated 

 ring by the young Phyttodoce and others, in the shedding of the temporary bristles noticed by 

 Busch and Leuckart in the young of a JVerine, and by M. de Quatrefages in the young stages of 

 Hermetta. I think there can be no doubt that the remarkable tuft of cilia occurring in the 

 young CepJialothrix on each side of the snout, and which attains its full development after the 

 long anterior whip has ceased to be conspicuous, is connected homologically with the entrance to 

 the cephalic sacs in the Enopla, and the fissures of the Lineidce, as well as with the ciliated ring 

 of Phyttodoce above mentioned. It is an embryonic type of a structure which disappears entirely 

 in the adult. The delicacy of the young at the period of the full development of the eye- 

 specks is an interesting feature ; but it prevented my observing their growth into perfect 

 animals. 



Thus, so far as development goes, Cephalotlirix is nearly allied to the Enopla, especially to 

 Tetrastemma dorsalis, Nemertes carcinophila, and probably to others of the group not yet 

 investigated; while, in the structure of its digestive system, circulatory apparatus, and the 

 unarmed proboscis with its bridled sheath, it leans rather towards the Lineidce. Prof. Keferstein 

 in his proposed classification of the Order rightly places the genus in a special Family, called 

 Gymnocephalidce, the chief characteristics described by him being : — Absence of cephalic 

 fissures ; brain like that of Folia (AmpMporus), but the superior ganglion covers the inferior much 

 less, and is advanced in front of it. He bases his statement of the relationship to the Enopla, as 

 it appears to me, on somewhat questionable grounds, for the ganglia are by no means closely 

 allied in form and structure to those of that group. 



III. — Reproduction of Lost Parts. 



In the Nemerteans, as in the Annelida proper, the reproduction of rejected parts and the repair 

 of wounds take place with accuracy and considerable rapidity. If but a fragment is left behind 

 the head, a new body and tail are reproduced in the majority. The severed posterior half 

 of the animal, or other headless fragment, seldom perfects a head in confinement, but 

 remains alive for a year or more, slowly turning round when irritated, and, moreover, developing 

 the generative products in its interior. Thus a specimen of Linens marinus, sent from St. 

 Andrews in September, broke into pieces on the journey ; yet six months afterwards most of the 

 fragments were alive, although the sea-water had not been changed more than once. The 

 head and anterior portion of the worm, which at first scarcely measured two inches, 

 had now grown a body and tail that when progressing extended at least seven inches, 

 and of course was capable of much greater elongation, so that it looked like an inde- 

 pendent animal ; and this was accomplished without the aid of any food, except perhaps what it 

 might have acquired from the fragments of its own body in the neighbourhood. Some of the 

 latter measured about a foot in length, and all were coiled in various ways, with the ends puckered, 

 and in most cases fixed by a whitish cicatrix, which was firmer at one end than the other, and 



