STRUCTURE AND DEVELOPMENT OF LOXOSOMA. 
315 
procta. The most important feature of Cyphonautes is the 
fact that the alimentary canal is well developed in the larva ; 
broadly speaking, it has the same arrangement as that of the 
Entoprocta, and is obviously functional during larval life, 
as can be concluded from the presence of food particles in the 
stomach, as indicated by Repiachoff (27). In the paper just 
referred to, Repiachoff has given figures of Cyphonautes, 
which I have been able to understand, partially at any rate, 
by means of a short description in the ‘ Zool. Anzeiger ’ (22) 
(the longer paper is in Russian). It i3 pointed out in this 
paper that the “ bud ” described by Hatschek (14) in 
Cyphonautes consists of two parts: 1 (i) an invagination, 
composed of columnar cells opening into the anterior side of the 
vestibule at its extreme ventral edge (fig. 24, zc ) ; this portion 
is surrounded by a ring of cilia ; (ii) of a mass of cells, cor- 
responding to the “ Entodermknospe ” described by Hatschek 
in Pedicellina, situated on the dorsal side of the invaginated 
part, and apparently (from another figure of RepiachofFs) 
giving rise to the first polypide of the colony. It appears to 
me not impossible, from this description, that the first part, 
the invaginated sac, corresponds to the epiblastic invagination 
which forms the dorsal organ of Loxosoma or Pedicellina. 
Ey reflecting the anterior portion of the body of an Ento- 
proctan larva into the vestibule, as Hatschek has suggested 
(although with a different identification of the dorsal organ), 
the aperture of the invagination might come to lie within the 
area of the latter, and the involuted sac (zc, fig. 24) may thus 
possibly represent the dorsal organ, which, instead of develop- 
ing to a functional ganglion, remains in its embryonic con- 
dition as a mere rudiment. The mass of cells between the sac 
zc in Cyphonautes and the sucker on the dorsal side of the 
larva, appears from Repiachoff’ s fig. 2 to be concerned in the 
larval budding; and it is important to notice that in the figure 
it is in intimate connection with the epiblast of the anterior 
ventral portion of the vestibule, but not with that forming the 
1 See fig. 24, in which are reproduced the essential details of Repiachoff’s 
Taf. i, fig. 1. 
