54 BULLETIN : MUSEUM OF COMPARATIVE ZOOLOGY. 



parallel and perpendicular to the longitudinal axis, though less regular 

 than the oblique ones, are easily recognizable. Owing to the surface of 

 the bell being curved, the arrangement in quincunx is not as evident as 

 I have represented it diagrammatically in Figure 20, but has more the 

 appearance shown in Figure 19, where the oblique and transverse rows 

 are somewhat curved, and the longitudinal ones diverge slightly as they 

 pass backwards. 



The outer layer (Fig. 7, st.'") is composed of a series of ridges, or, 

 more properly speaking, plates. These plates, cut crosswise, are shown 

 in Figure 16 {st.'"), which represents a transverse section through the 

 middle of the bell-shaped ciliated portion of the body, about in the place 

 indicated by the line 16-16 in Figure 15. The plates are seen more 

 clearly in the diagrammatic cross section shown in Figure 17, st.'". 

 In surface views of the animal each of the plates of the nipple appears 

 to be split into two at the region of the fissure (Fig. \\,fis.); this, 

 however, is only an optical illusion. Although the plates of the bell- 

 shaped pai't are twice as numerous as those of the nipple-like part, 

 there is no continuity between the two sets, the appearance of splitting 

 being due to the overlapping of the anterior set over the posterior one. 

 The nature of this overlapping (Fig. \2, fis.) may be seen in optical 

 longitudinal sections. The plates of the outer layer coincide with the 

 longitudinal rows of cilia. Their inner edges are slightly serrated 

 (Fig. 8, csp.). owing to a slight inwai'd prolongation of the plate wher- 

 ever a cilium in passing through meets it. It is this condition which 

 causes the interrupted appearance of the plates seen in tangential sec- 

 tions (Fig. 20, c.<t/>.). 



Tlie bell-shaped region is generally sharply marked off from the an- 

 terior or nipple-like part by a deep constriction (Plate 1, Figs. 1-6) ; 

 but sometimes there is a more gradual transition (Plate 2, Figs. 7, 8). 

 The boundary between the " bell " and the " body " is ordinarily 

 marked by a deep constriction (Plate 1, Figs. 1, 6), or a shoulder-like 

 projection of the "body" beyond the outline of the "bell" (Plate 2, 

 Figs. 7, 15). The place of this constriction may for convenience be 

 called the lip of the bell, though the pendant central mass of the proto- 

 plasm of the bell containing the nucleus often projects far behind this 

 rim or lip. When the animal is at rest, the bell is of fairly regular and 

 symmetrical form (Figs. 13, 15), but the extreme mobility of the whole 

 head region causes it when in action to take a great variety of shapes 

 (Plate 1, Figs. 1-6). 



The protoplasm of the bell is differentiated into two distinct kinds, 



